From d2ebb4103ff349af6dac14955bf93e57487a6694 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Wed, 30 Apr 2008 17:56:04 +0200 Subject: KVM: SVM: add tracing support for TDP page faults To distinguish between real page faults and nested page faults they should be traced as different events. This is implemented by this patch. Signed-off-by: Joerg Roedel Signed-off-by: Avi Kivity --- include/asm-x86/kvm.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/asm-x86/kvm.h b/include/asm-x86/kvm.h index 80eefef2cc76..6f1840812e59 100644 --- a/include/asm-x86/kvm.h +++ b/include/asm-x86/kvm.h @@ -228,5 +228,6 @@ struct kvm_pit_state { #define KVM_TRC_CLTS (KVM_TRC_HANDLER + 0x12) #define KVM_TRC_LMSW (KVM_TRC_HANDLER + 0x13) #define KVM_TRC_APIC_ACCESS (KVM_TRC_HANDLER + 0x14) +#define KVM_TRC_TDP_FAULT (KVM_TRC_HANDLER + 0x15) #endif -- cgit From 1b7fcd3263e5f12dba43d27b64e1578bec070c28 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 15 May 2008 13:51:35 +0300 Subject: KVM: MMU: Fix false flooding when a pte points to page table The KVM MMU tries to detect when a speculative pte update is not actually used by demand fault, by checking the accessed bit of the shadow pte. If the shadow pte has not been accessed, we deem that page table flooded and remove the shadow page table, allowing further pte updates to proceed without emulation. However, if the pte itself points at a page table and only used for write operations, the accessed bit will never be set since all access will happen through the emulator. This is exactly what happens with kscand on old (2.4.x) HIGHMEM kernels. The kernel points a kmap_atomic() pte at a page table, and then proceeds with read-modify-write operations to look at the dirty and accessed bits. We get a false flood trigger on the kmap ptes, which results in the mmu spending all its time setting up and tearing down shadows. Fix by setting the shadow accessed bit on emulated accesses. Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 17 ++++++++++++++++- arch/x86/kvm/mmu.h | 3 ++- include/asm-x86/kvm_host.h | 1 + 3 files changed, 19 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 8e449dbcc596..53f1ed852ca2 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -1122,8 +1122,10 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, else kvm_release_pfn_clean(pfn); } - if (!ptwrite || !*ptwrite) + if (speculative) { vcpu->arch.last_pte_updated = shadow_pte; + vcpu->arch.last_pte_gfn = gfn; + } } static void nonpaging_new_cr3(struct kvm_vcpu *vcpu) @@ -1671,6 +1673,18 @@ static void mmu_guess_page_from_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, vcpu->arch.update_pte.pfn = pfn; } +static void kvm_mmu_access_page(struct kvm_vcpu *vcpu, gfn_t gfn) +{ + u64 *spte = vcpu->arch.last_pte_updated; + + if (spte + && vcpu->arch.last_pte_gfn == gfn + && shadow_accessed_mask + && !(*spte & shadow_accessed_mask) + && is_shadow_present_pte(*spte)) + set_bit(PT_ACCESSED_SHIFT, (unsigned long *)spte); +} + void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, const u8 *new, int bytes) { @@ -1694,6 +1708,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, pgprintk("%s: gpa %llx bytes %d\n", __func__, gpa, bytes); mmu_guess_page_from_pte_write(vcpu, gpa, new, bytes); spin_lock(&vcpu->kvm->mmu_lock); + kvm_mmu_access_page(vcpu, gfn); kvm_mmu_free_some_pages(vcpu); ++vcpu->kvm->stat.mmu_pte_write; kvm_mmu_audit(vcpu, "pre pte write"); diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index 1730757bbc7a..258e5d56298e 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -15,7 +15,8 @@ #define PT_USER_MASK (1ULL << 2) #define PT_PWT_MASK (1ULL << 3) #define PT_PCD_MASK (1ULL << 4) -#define PT_ACCESSED_MASK (1ULL << 5) +#define PT_ACCESSED_SHIFT 5 +#define PT_ACCESSED_MASK (1ULL << PT_ACCESSED_SHIFT) #define PT_DIRTY_MASK (1ULL << 6) #define PT_PAGE_SIZE_MASK (1ULL << 7) #define PT_PAT_MASK (1ULL << 7) diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 844f2a89afbc..c2d066e185f4 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -243,6 +243,7 @@ struct kvm_vcpu_arch { gfn_t last_pt_write_gfn; int last_pt_write_count; u64 *last_pte_updated; + gfn_t last_pte_gfn; struct { gfn_t gfn; /* presumed gfn during guest pte update */ -- cgit From 4ecac3fd6dc2629ad76a658a486f081c44aef10e Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 13 May 2008 13:23:38 +0300 Subject: KVM: Handle virtualization instruction #UD faults during reboot KVM turns off hardware virtualization extensions during reboot, in order to disassociate the memory used by the virtualization extensions from the processor, and in order to have the system in a consistent state. Unfortunately virtual machines may still be running while this goes on, and once virtualization extensions are turned off, any virtulization instruction will #UD on execution. Fix by adding an exception handler to virtualization instructions; if we get an exception during reboot, we simply spin waiting for the reset to complete. If it's a true exception, BUG() so we can have our stack trace. Signed-off-by: Avi Kivity --- arch/x86/kvm/svm.c | 20 +++++++++++--------- arch/x86/kvm/vmx.c | 25 ++++++++++++++----------- include/asm-x86/kvm_host.h | 24 ++++++++++++++++++++++++ virt/kvm/kvm_main.c | 15 +++++++++++++++ 4 files changed, 64 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 992ab7115871..9390a31c06f4 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -27,6 +27,8 @@ #include +#define __ex(x) __kvm_handle_fault_on_reboot(x) + MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); @@ -129,17 +131,17 @@ static inline void push_irq(struct kvm_vcpu *vcpu, u8 irq) static inline void clgi(void) { - asm volatile (SVM_CLGI); + asm volatile (__ex(SVM_CLGI)); } static inline void stgi(void) { - asm volatile (SVM_STGI); + asm volatile (__ex(SVM_STGI)); } static inline void invlpga(unsigned long addr, u32 asid) { - asm volatile (SVM_INVLPGA :: "a"(addr), "c"(asid)); + asm volatile (__ex(SVM_INVLPGA) :: "a"(addr), "c"(asid)); } static inline unsigned long kvm_read_cr2(void) @@ -1758,17 +1760,17 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) /* Enter guest mode */ "push %%rax \n\t" "mov %c[vmcb](%[svm]), %%rax \n\t" - SVM_VMLOAD "\n\t" - SVM_VMRUN "\n\t" - SVM_VMSAVE "\n\t" + __ex(SVM_VMLOAD) "\n\t" + __ex(SVM_VMRUN) "\n\t" + __ex(SVM_VMSAVE) "\n\t" "pop %%rax \n\t" #else /* Enter guest mode */ "push %%eax \n\t" "mov %c[vmcb](%[svm]), %%eax \n\t" - SVM_VMLOAD "\n\t" - SVM_VMRUN "\n\t" - SVM_VMSAVE "\n\t" + __ex(SVM_VMLOAD) "\n\t" + __ex(SVM_VMRUN) "\n\t" + __ex(SVM_VMSAVE) "\n\t" "pop %%eax \n\t" #endif diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index fff8e23433d6..b80b4d141637 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -30,6 +30,8 @@ #include #include +#define __ex(x) __kvm_handle_fault_on_reboot(x) + MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); @@ -278,7 +280,7 @@ static inline void __invvpid(int ext, u16 vpid, gva_t gva) u64 gva; } operand = { vpid, 0, gva }; - asm volatile (ASM_VMX_INVVPID + asm volatile (__ex(ASM_VMX_INVVPID) /* CF==1 or ZF==1 --> rc = -1 */ "; ja 1f ; ud2 ; 1:" : : "a"(&operand), "c"(ext) : "cc", "memory"); @@ -290,7 +292,7 @@ static inline void __invept(int ext, u64 eptp, gpa_t gpa) u64 eptp, gpa; } operand = {eptp, gpa}; - asm volatile (ASM_VMX_INVEPT + asm volatile (__ex(ASM_VMX_INVEPT) /* CF==1 or ZF==1 --> rc = -1 */ "; ja 1f ; ud2 ; 1:\n" : : "a" (&operand), "c" (ext) : "cc", "memory"); @@ -311,7 +313,7 @@ static void vmcs_clear(struct vmcs *vmcs) u64 phys_addr = __pa(vmcs); u8 error; - asm volatile (ASM_VMX_VMCLEAR_RAX "; setna %0" + asm volatile (__ex(ASM_VMX_VMCLEAR_RAX) "; setna %0" : "=g"(error) : "a"(&phys_addr), "m"(phys_addr) : "cc", "memory"); if (error) @@ -378,7 +380,7 @@ static unsigned long vmcs_readl(unsigned long field) { unsigned long value; - asm volatile (ASM_VMX_VMREAD_RDX_RAX + asm volatile (__ex(ASM_VMX_VMREAD_RDX_RAX) : "=a"(value) : "d"(field) : "cc"); return value; } @@ -413,7 +415,7 @@ static void vmcs_writel(unsigned long field, unsigned long value) { u8 error; - asm volatile (ASM_VMX_VMWRITE_RAX_RDX "; setna %0" + asm volatile (__ex(ASM_VMX_VMWRITE_RAX_RDX) "; setna %0" : "=q"(error) : "a"(value), "d"(field) : "cc"); if (unlikely(error)) vmwrite_error(field, value); @@ -621,7 +623,7 @@ static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu) u8 error; per_cpu(current_vmcs, cpu) = vmx->vmcs; - asm volatile (ASM_VMX_VMPTRLD_RAX "; setna %0" + asm volatile (__ex(ASM_VMX_VMPTRLD_RAX) "; setna %0" : "=g"(error) : "a"(&phys_addr), "m"(phys_addr) : "cc"); if (error) @@ -1030,13 +1032,14 @@ static void hardware_enable(void *garbage) MSR_IA32_FEATURE_CONTROL_LOCKED | MSR_IA32_FEATURE_CONTROL_VMXON_ENABLED); write_cr4(read_cr4() | X86_CR4_VMXE); /* FIXME: not cpu hotplug safe */ - asm volatile (ASM_VMX_VMXON_RAX : : "a"(&phys_addr), "m"(phys_addr) + asm volatile (ASM_VMX_VMXON_RAX + : : "a"(&phys_addr), "m"(phys_addr) : "memory", "cc"); } static void hardware_disable(void *garbage) { - asm volatile (ASM_VMX_VMXOFF : : : "cc"); + asm volatile (__ex(ASM_VMX_VMXOFF) : : : "cc"); write_cr4(read_cr4() & ~X86_CR4_VMXE); } @@ -2834,7 +2837,7 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) "push %%edx; push %%ebp;" "push %%ecx \n\t" #endif - ASM_VMX_VMWRITE_RSP_RDX "\n\t" + __ex(ASM_VMX_VMWRITE_RSP_RDX) "\n\t" /* Check if vmlaunch of vmresume is needed */ "cmpl $0, %c[launched](%0) \n\t" /* Load guest registers. Don't clobber flags. */ @@ -2869,9 +2872,9 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) #endif /* Enter guest mode */ "jne .Llaunched \n\t" - ASM_VMX_VMLAUNCH "\n\t" + __ex(ASM_VMX_VMLAUNCH) "\n\t" "jmp .Lkvm_vmx_return \n\t" - ".Llaunched: " ASM_VMX_VMRESUME "\n\t" + ".Llaunched: " __ex(ASM_VMX_VMRESUME) "\n\t" ".Lkvm_vmx_return: " /* Save guest registers, load host registers, keep flags */ #ifdef CONFIG_X86_64 diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index c2d066e185f4..0df9d5fa281a 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -692,4 +692,28 @@ enum { trace_mark(kvm_trace_##name, "%u %p %u %u %u %u %u %u", KVM_TRC_##evt, \ vcpu, 0, 0, 0, 0, 0, 0) +#ifdef CONFIG_64BIT +#define KVM_EX_ENTRY ".quad" +#else +#define KVM_EX_ENTRY ".long" +#endif + +/* + * Hardware virtualization extension instructions may fault if a + * reboot turns off virtualization while processes are running. + * Trap the fault and ignore the instruction if that happens. + */ +asmlinkage void kvm_handle_fault_on_reboot(void); + +#define __kvm_handle_fault_on_reboot(insn) \ + "666: " insn "\n\t" \ + ".pushsection .text.fixup, \"ax\" \n" \ + "667: \n\t" \ + "push $666b \n\t" \ + "jmp kvm_handle_fault_on_reboot \n\t" \ + ".popsection \n\t" \ + ".pushsection __ex_table, \"a\" \n\t" \ + KVM_EX_ENTRY " 666b, 667b \n\t" \ + ".popsection" + #endif diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index f9dd20606c40..e4bf88a9ee4e 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -65,6 +65,8 @@ struct dentry *kvm_debugfs_dir; static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl, unsigned long arg); +bool kvm_rebooting; + static inline int valid_vcpu(int n) { return likely(n >= 0 && n < KVM_MAX_VCPUS); @@ -1301,6 +1303,18 @@ static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val, return NOTIFY_OK; } + +asmlinkage void kvm_handle_fault_on_reboot(void) +{ + if (kvm_rebooting) + /* spin while reset goes on */ + while (true) + ; + /* Fault while not rebooting. We want the trace. */ + BUG(); +} +EXPORT_SYMBOL_GPL(kvm_handle_fault_on_reboot); + static int kvm_reboot(struct notifier_block *notifier, unsigned long val, void *v) { @@ -1310,6 +1324,7 @@ static int kvm_reboot(struct notifier_block *notifier, unsigned long val, * in vmx root mode. */ printk(KERN_INFO "kvm: exiting hardware virtualization\n"); + kvm_rebooting = true; on_each_cpu(hardware_disable, NULL, 1); } return NOTIFY_OK; -- cgit From 7cc8883074b040aa8c1ebd3a17463b0ea3a9ef16 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 13 May 2008 16:29:20 +0300 Subject: KVM: Remove decache_vcpus_on_cpu() and related callbacks Obsoleted by the vmx-specific per-cpu list. Signed-off-by: Avi Kivity --- arch/ia64/kvm/kvm-ia64.c | 8 -------- arch/powerpc/kvm/powerpc.c | 4 ---- arch/s390/kvm/kvm-s390.c | 4 ---- arch/x86/kvm/svm.c | 5 ----- arch/x86/kvm/vmx.c | 6 ------ arch/x86/kvm/x86.c | 8 -------- include/asm-x86/kvm_host.h | 1 - include/linux/kvm_host.h | 3 --- virt/kvm/kvm_main.c | 1 - 9 files changed, 40 deletions(-) (limited to 'include') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 68c978be9a51..7c504be57972 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1035,14 +1035,6 @@ static void kvm_free_vmm_area(void) } } -/* - * Make sure that a cpu that is being hot-unplugged does not have any vcpus - * cached on it. Leave it as blank for IA64. - */ -void decache_vcpus_on_cpu(int cpu) -{ -} - static void vti_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { } diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 777e0f34e0ea..0513b359851b 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -240,10 +240,6 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) { } -void decache_vcpus_on_cpu(int cpu) -{ -} - int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) { diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 6558b09ff579..4585c8ac2b0c 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -79,10 +79,6 @@ void kvm_arch_hardware_disable(void *garbage) { } -void decache_vcpus_on_cpu(int cpu) -{ -} - int kvm_arch_hardware_setup(void) { return 0; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 9390a31c06f4..238e8f3afaf4 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -709,10 +709,6 @@ static void svm_vcpu_put(struct kvm_vcpu *vcpu) rdtscll(vcpu->arch.host_tsc); } -static void svm_vcpu_decache(struct kvm_vcpu *vcpu) -{ -} - static void svm_cache_regs(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1933,7 +1929,6 @@ static struct kvm_x86_ops svm_x86_ops = { .prepare_guest_switch = svm_prepare_guest_switch, .vcpu_load = svm_vcpu_load, .vcpu_put = svm_vcpu_put, - .vcpu_decache = svm_vcpu_decache, .set_guest_debug = svm_guest_debug, .get_msr = svm_get_msr, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 4d179d106376..b99bb37e5dec 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -692,11 +692,6 @@ static void vmx_fpu_deactivate(struct kvm_vcpu *vcpu) update_exception_bitmap(vcpu); } -static void vmx_vcpu_decache(struct kvm_vcpu *vcpu) -{ - vcpu_clear(to_vmx(vcpu)); -} - static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu) { return vmcs_readl(GUEST_RFLAGS); @@ -3114,7 +3109,6 @@ static struct kvm_x86_ops vmx_x86_ops = { .prepare_guest_switch = vmx_save_host_state, .vcpu_load = vmx_vcpu_load, .vcpu_put = vmx_vcpu_put, - .vcpu_decache = vmx_vcpu_decache, .set_guest_debug = set_guest_debug, .guest_debug_pre = kvm_guest_debug_pre, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 8c14ddcaba70..fd03b4465bcc 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -817,14 +817,6 @@ out: return r; } -/* - * Make sure that a cpu that is being hot-unplugged does not have any vcpus - * cached on it. - */ -void decache_vcpus_on_cpu(int cpu) -{ -} - int kvm_dev_ioctl_check_extension(long ext) { int r; diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 0df9d5fa281a..4bcdc7de07b5 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -380,7 +380,6 @@ struct kvm_x86_ops { void (*prepare_guest_switch)(struct kvm_vcpu *vcpu); void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu); void (*vcpu_put)(struct kvm_vcpu *vcpu); - void (*vcpu_decache)(struct kvm_vcpu *vcpu); int (*set_guest_debug)(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index de9d1df4bba2..865dcbcb891f 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -135,9 +135,6 @@ void kvm_vcpu_uninit(struct kvm_vcpu *vcpu); void vcpu_load(struct kvm_vcpu *vcpu); void vcpu_put(struct kvm_vcpu *vcpu); -void decache_vcpus_on_cpu(int cpu); - - int kvm_init(void *opaque, unsigned int vcpu_size, struct module *module); void kvm_exit(void); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index e4bf88a9ee4e..83a0e5ce6037 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1273,7 +1273,6 @@ static void hardware_disable(void *junk) if (!cpu_isset(cpu, cpus_hardware_enabled)) return; cpu_clear(cpu, cpus_hardware_enabled); - decache_vcpus_on_cpu(cpu); kvm_arch_hardware_disable(NULL); } -- cgit From 3419ffc8e45a5344abc87684cbca6cdc5c9c8a01 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Thu, 15 May 2008 09:52:48 +0800 Subject: KVM: IOAPIC/LAPIC: Enable NMI support [avi: fix ia64 build breakage] Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/lapic.c | 3 ++- arch/x86/kvm/x86.c | 6 ++++++ include/asm-ia64/kvm_host.h | 2 ++ include/asm-x86/kvm_host.h | 4 ++++ virt/kvm/ioapic.c | 20 ++++++++++++++++++-- 5 files changed, 32 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index f9201fbc61d1..e48d19394031 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -356,8 +356,9 @@ static int __apic_accept_irq(struct kvm_lapic *apic, int delivery_mode, case APIC_DM_SMI: printk(KERN_DEBUG "Ignoring guest SMI\n"); break; + case APIC_DM_NMI: - printk(KERN_DEBUG "Ignoring guest NMI\n"); + kvm_inject_nmi(vcpu); break; case APIC_DM_INIT: diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 5f00c60f0aff..19974dde6567 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -173,6 +173,12 @@ void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long addr, kvm_queue_exception_e(vcpu, PF_VECTOR, error_code); } +void kvm_inject_nmi(struct kvm_vcpu *vcpu) +{ + vcpu->arch.nmi_pending = 1; +} +EXPORT_SYMBOL_GPL(kvm_inject_nmi); + void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code) { WARN_ON(vcpu->arch.exception.pending); diff --git a/include/asm-ia64/kvm_host.h b/include/asm-ia64/kvm_host.h index c082c208c1f3..5c958b0c46b1 100644 --- a/include/asm-ia64/kvm_host.h +++ b/include/asm-ia64/kvm_host.h @@ -521,4 +521,6 @@ int kvm_emulate_halt(struct kvm_vcpu *vcpu); int kvm_pal_emul(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run); void kvm_sal_emul(struct kvm_vcpu *vcpu); +static inline void kvm_inject_nmi(struct kvm_vcpu *vcpu) {} + #endif diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 4bcdc7de07b5..b66621935eb7 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -288,6 +288,8 @@ struct kvm_vcpu_arch { unsigned int hv_clock_tsc_khz; unsigned int time_offset; struct page *time_page; + + bool nmi_pending; }; struct kvm_mem_alias { @@ -515,6 +517,8 @@ void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code); void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long cr2, u32 error_code); +void kvm_inject_nmi(struct kvm_vcpu *vcpu); + void fx_init(struct kvm_vcpu *vcpu); int emulator_read_std(unsigned long addr, diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index 44589088941f..d0c668c6959e 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -146,6 +146,11 @@ static int ioapic_inj_irq(struct kvm_ioapic *ioapic, return kvm_apic_set_irq(vcpu, vector, trig_mode); } +static void ioapic_inj_nmi(struct kvm_vcpu *vcpu) +{ + kvm_inject_nmi(vcpu); +} + static u32 ioapic_get_delivery_bitmask(struct kvm_ioapic *ioapic, u8 dest, u8 dest_mode) { @@ -239,8 +244,19 @@ static int ioapic_deliver(struct kvm_ioapic *ioapic, int irq) } } break; - - /* TODO: NMI */ + case IOAPIC_NMI: + for (vcpu_id = 0; deliver_bitmask != 0; vcpu_id++) { + if (!(deliver_bitmask & (1 << vcpu_id))) + continue; + deliver_bitmask &= ~(1 << vcpu_id); + vcpu = ioapic->kvm->vcpus[vcpu_id]; + if (vcpu) + ioapic_inj_nmi(vcpu); + else + ioapic_debug("NMI to vcpu %d failed\n", + vcpu->vcpu_id); + } + break; default: printk(KERN_WARNING "Unsupported delivery mode %d\n", delivery_mode); -- cgit From f08864b42a45581a64558aa5b6b673c77b97ee5d Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Thu, 15 May 2008 18:23:25 +0800 Subject: KVM: VMX: Enable NMI with in-kernel irqchip Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.c | 124 ++++++++++++++++++++++++++++++++++++++------- arch/x86/kvm/vmx.h | 12 ++++- arch/x86/kvm/x86.c | 1 + include/asm-x86/kvm_host.h | 1 + 4 files changed, 119 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index b99bb37e5dec..1bb994657208 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -264,6 +264,11 @@ static inline int cpu_has_vmx_vpid(void) SECONDARY_EXEC_ENABLE_VPID); } +static inline int cpu_has_virtual_nmis(void) +{ + return vmcs_config.pin_based_exec_ctrl & PIN_BASED_VIRTUAL_NMIS; +} + static int __find_msr_index(struct vcpu_vmx *vmx, u32 msr) { int i; @@ -1088,7 +1093,7 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) u32 _vmentry_control = 0; min = PIN_BASED_EXT_INTR_MASK | PIN_BASED_NMI_EXITING; - opt = 0; + opt = PIN_BASED_VIRTUAL_NMIS; if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PINBASED_CTLS, &_pin_based_exec_control) < 0) return -EIO; @@ -2130,6 +2135,13 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq) irq | INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK); } +static void vmx_inject_nmi(struct kvm_vcpu *vcpu) +{ + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + INTR_TYPE_NMI_INTR | INTR_INFO_VALID_MASK | NMI_VECTOR); + vcpu->arch.nmi_pending = 0; +} + static void kvm_do_inject_irq(struct kvm_vcpu *vcpu) { int word_index = __ffs(vcpu->arch.irq_summary); @@ -2653,6 +2665,19 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 1; } +static int handle_nmi_window(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 cpu_based_vm_exec_control; + + /* clear pending NMI */ + cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); + cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_NMI_PENDING; + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); + ++vcpu->stat.nmi_window_exits; + + return 1; +} + /* * The exit handlers return 1 if the exit was handled fully and guest execution * may resume. Otherwise they set the kvm_run parameter to indicate what needs @@ -2663,6 +2688,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu, [EXIT_REASON_EXCEPTION_NMI] = handle_exception, [EXIT_REASON_EXTERNAL_INTERRUPT] = handle_external_interrupt, [EXIT_REASON_TRIPLE_FAULT] = handle_triple_fault, + [EXIT_REASON_NMI_WINDOW] = handle_nmi_window, [EXIT_REASON_IO_INSTRUCTION] = handle_io, [EXIT_REASON_CR_ACCESS] = handle_cr, [EXIT_REASON_DR_ACCESS] = handle_dr, @@ -2750,17 +2776,52 @@ static void enable_irq_window(struct kvm_vcpu *vcpu) vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); } +static void enable_nmi_window(struct kvm_vcpu *vcpu) +{ + u32 cpu_based_vm_exec_control; + + if (!cpu_has_virtual_nmis()) + return; + + cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); + cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_NMI_PENDING; + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); +} + +static int vmx_nmi_enabled(struct kvm_vcpu *vcpu) +{ + u32 guest_intr = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO); + return !(guest_intr & (GUEST_INTR_STATE_NMI | + GUEST_INTR_STATE_MOV_SS | + GUEST_INTR_STATE_STI)); +} + +static int vmx_irq_enabled(struct kvm_vcpu *vcpu) +{ + u32 guest_intr = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO); + return (!(guest_intr & (GUEST_INTR_STATE_MOV_SS | + GUEST_INTR_STATE_STI)) && + (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF)); +} + +static void enable_intr_window(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.nmi_pending) + enable_nmi_window(vcpu); + else if (kvm_cpu_has_interrupt(vcpu)) + enable_irq_window(vcpu); +} + static void vmx_intr_assist(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); - u32 idtv_info_field, intr_info_field; - int has_ext_irq, interrupt_window_open; + u32 idtv_info_field, intr_info_field, exit_intr_info_field; int vector; update_tpr_threshold(vcpu); - has_ext_irq = kvm_cpu_has_interrupt(vcpu); intr_info_field = vmcs_read32(VM_ENTRY_INTR_INFO_FIELD); + exit_intr_info_field = vmcs_read32(VM_EXIT_INTR_INFO); idtv_info_field = vmx->idt_vectoring_info; if (intr_info_field & INTR_INFO_VALID_MASK) { if (idtv_info_field & INTR_INFO_VALID_MASK) { @@ -2768,8 +2829,7 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) if (printk_ratelimit()) printk(KERN_ERR "Fault when IDT_Vectoring\n"); } - if (has_ext_irq) - enable_irq_window(vcpu); + enable_intr_window(vcpu); return; } if (unlikely(idtv_info_field & INTR_INFO_VALID_MASK)) { @@ -2779,30 +2839,56 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) u8 vect = idtv_info_field & VECTORING_INFO_VECTOR_MASK; vmx_inject_irq(vcpu, vect); - if (unlikely(has_ext_irq)) - enable_irq_window(vcpu); + enable_intr_window(vcpu); return; } KVMTRACE_1D(REDELIVER_EVT, vcpu, idtv_info_field, handler); - vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field); + /* + * SDM 3: 25.7.1.2 + * Clear bit "block by NMI" before VM entry if a NMI delivery + * faulted. + */ + if ((idtv_info_field & VECTORING_INFO_TYPE_MASK) + == INTR_TYPE_NMI_INTR && cpu_has_virtual_nmis()) + vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, + vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & + ~GUEST_INTR_STATE_NMI); + + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field + & ~INTR_INFO_RESVD_BITS_MASK); vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, vmcs_read32(VM_EXIT_INSTRUCTION_LEN)); if (unlikely(idtv_info_field & INTR_INFO_DELIVER_CODE_MASK)) vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, vmcs_read32(IDT_VECTORING_ERROR_CODE)); - if (unlikely(has_ext_irq)) - enable_irq_window(vcpu); + enable_intr_window(vcpu); return; } - if (!has_ext_irq) + if (cpu_has_virtual_nmis()) { + /* + * SDM 3: 25.7.1.2 + * Re-set bit "block by NMI" before VM entry if vmexit caused by + * a guest IRET fault. + */ + if ((exit_intr_info_field & INTR_INFO_UNBLOCK_NMI) && + (exit_intr_info_field & INTR_INFO_VECTOR_MASK) != 8) + vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, + vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) | + GUEST_INTR_STATE_NMI); + else if (vcpu->arch.nmi_pending) { + if (vmx_nmi_enabled(vcpu)) + vmx_inject_nmi(vcpu); + enable_intr_window(vcpu); + return; + } + + } + if (!kvm_cpu_has_interrupt(vcpu)) return; - interrupt_window_open = - ((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) && - (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0); - if (interrupt_window_open) { + if (vmx_irq_enabled(vcpu)) { vector = kvm_cpu_get_interrupt(vcpu); vmx_inject_irq(vcpu, vector); kvm_timer_intr_post(vcpu, vector); @@ -2963,7 +3049,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) fixup_rmode_irq(vmx); vcpu->arch.interrupt_window_open = - (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0; + (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & + (GUEST_INTR_STATE_STI | GUEST_INTR_STATE_MOV_SS)) == 0; asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS)); vmx->launched = 1; @@ -2971,7 +3058,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) intr_info = vmcs_read32(VM_EXIT_INTR_INFO); /* We need to handle NMIs before interrupts are enabled */ - if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) { /* nmi */ + if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200 && + (intr_info & INTR_INFO_VALID_MASK)) { KVMTRACE_0D(NMI, vcpu, handler); asm("int $2"); } diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h index 79d94c610dfe..425a13436b3f 100644 --- a/arch/x86/kvm/vmx.h +++ b/arch/x86/kvm/vmx.h @@ -40,6 +40,7 @@ #define CPU_BASED_CR8_LOAD_EXITING 0x00080000 #define CPU_BASED_CR8_STORE_EXITING 0x00100000 #define CPU_BASED_TPR_SHADOW 0x00200000 +#define CPU_BASED_VIRTUAL_NMI_PENDING 0x00400000 #define CPU_BASED_MOV_DR_EXITING 0x00800000 #define CPU_BASED_UNCOND_IO_EXITING 0x01000000 #define CPU_BASED_USE_IO_BITMAPS 0x02000000 @@ -216,7 +217,7 @@ enum vmcs_field { #define EXIT_REASON_TRIPLE_FAULT 2 #define EXIT_REASON_PENDING_INTERRUPT 7 - +#define EXIT_REASON_NMI_WINDOW 8 #define EXIT_REASON_TASK_SWITCH 9 #define EXIT_REASON_CPUID 10 #define EXIT_REASON_HLT 12 @@ -251,7 +252,9 @@ enum vmcs_field { #define INTR_INFO_VECTOR_MASK 0xff /* 7:0 */ #define INTR_INFO_INTR_TYPE_MASK 0x700 /* 10:8 */ #define INTR_INFO_DELIVER_CODE_MASK 0x800 /* 11 */ +#define INTR_INFO_UNBLOCK_NMI 0x1000 /* 12 */ #define INTR_INFO_VALID_MASK 0x80000000 /* 31 */ +#define INTR_INFO_RESVD_BITS_MASK 0x7ffff000 #define VECTORING_INFO_VECTOR_MASK INTR_INFO_VECTOR_MASK #define VECTORING_INFO_TYPE_MASK INTR_INFO_INTR_TYPE_MASK @@ -259,9 +262,16 @@ enum vmcs_field { #define VECTORING_INFO_VALID_MASK INTR_INFO_VALID_MASK #define INTR_TYPE_EXT_INTR (0 << 8) /* external interrupt */ +#define INTR_TYPE_NMI_INTR (2 << 8) /* NMI */ #define INTR_TYPE_EXCEPTION (3 << 8) /* processor exception */ #define INTR_TYPE_SOFT_INTR (4 << 8) /* software interrupt */ +/* GUEST_INTERRUPTIBILITY_INFO flags. */ +#define GUEST_INTR_STATE_STI 0x00000001 +#define GUEST_INTR_STATE_MOV_SS 0x00000002 +#define GUEST_INTR_STATE_SMI 0x00000004 +#define GUEST_INTR_STATE_NMI 0x00000008 + /* * Exit Qualifications for MOV for Control Register Access */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 19974dde6567..05b54976c891 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -72,6 +72,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "mmio_exits", VCPU_STAT(mmio_exits) }, { "signal_exits", VCPU_STAT(signal_exits) }, { "irq_window", VCPU_STAT(irq_window_exits) }, + { "nmi_window", VCPU_STAT(nmi_window_exits) }, { "halt_exits", VCPU_STAT(halt_exits) }, { "halt_wakeup", VCPU_STAT(halt_wakeup) }, { "hypercalls", VCPU_STAT(hypercalls) }, diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index b66621935eb7..bacb1e24036e 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -347,6 +347,7 @@ struct kvm_vcpu_stat { u32 mmio_exits; u32 signal_exits; u32 irq_window_exits; + u32 nmi_window_exits; u32 halt_exits; u32 halt_wakeup; u32 request_irq_exits; -- cgit From 81609e3e26508840a1b51414376f2541dd191483 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 27 May 2008 16:26:01 +0300 Subject: KVM: Order segment register constants in the same way as cpu operand encoding This can be used to simplify the x86 instruction decoder. Signed-off-by: Avi Kivity --- include/asm-x86/kvm_host.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index bacb1e24036e..075598b4e3f3 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -109,12 +109,12 @@ enum { }; enum { + VCPU_SREG_ES, VCPU_SREG_CS, + VCPU_SREG_SS, VCPU_SREG_DS, - VCPU_SREG_ES, VCPU_SREG_FS, VCPU_SREG_GS, - VCPU_SREG_SS, VCPU_SREG_TR, VCPU_SREG_LDTR, }; -- cgit From 9ba075a664dff836fd6fb93f90fcc827f7683d91 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 26 May 2008 20:06:35 +0300 Subject: KVM: MTRR support Add emulation for the memory type range registers, needed by VMware esx 3.5, and by pci device assignment. Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 52 ++++++++++++++++++++++++++++++++++++++++++---- include/asm-x86/kvm_host.h | 3 +++ 2 files changed, 51 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 05b54976c891..5f67a7c54e82 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -611,6 +611,38 @@ static void kvm_write_guest_time(struct kvm_vcpu *v) mark_page_dirty(v->kvm, vcpu->time >> PAGE_SHIFT); } +static bool msr_mtrr_valid(unsigned msr) +{ + switch (msr) { + case 0x200 ... 0x200 + 2 * KVM_NR_VAR_MTRR - 1: + case MSR_MTRRfix64K_00000: + case MSR_MTRRfix16K_80000: + case MSR_MTRRfix16K_A0000: + case MSR_MTRRfix4K_C0000: + case MSR_MTRRfix4K_C8000: + case MSR_MTRRfix4K_D0000: + case MSR_MTRRfix4K_D8000: + case MSR_MTRRfix4K_E0000: + case MSR_MTRRfix4K_E8000: + case MSR_MTRRfix4K_F0000: + case MSR_MTRRfix4K_F8000: + case MSR_MTRRdefType: + case MSR_IA32_CR_PAT: + return true; + case 0x2f8: + return true; + } + return false; +} + +static int set_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 data) +{ + if (!msr_mtrr_valid(msr)) + return 1; + + vcpu->arch.mtrr[msr - 0x200] = data; + return 0; +} int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) { @@ -632,8 +664,9 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) break; case MSR_IA32_UCODE_REV: case MSR_IA32_UCODE_WRITE: - case 0x200 ... 0x2ff: /* MTRRs */ break; + case 0x200 ... 0x2ff: + return set_msr_mtrr(vcpu, msr, data); case MSR_IA32_APICBASE: kvm_set_apic_base(vcpu, data); break; @@ -691,6 +724,15 @@ int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) return kvm_x86_ops->get_msr(vcpu, msr_index, pdata); } +static int get_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) +{ + if (!msr_mtrr_valid(msr)) + return 1; + + *pdata = vcpu->arch.mtrr[msr - 0x200]; + return 0; +} + int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) { u64 data; @@ -712,11 +754,13 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_IA32_MC0_MISC+16: case MSR_IA32_UCODE_REV: case MSR_IA32_EBL_CR_POWERON: - /* MTRR registers */ - case 0xfe: - case 0x200 ... 0x2ff: data = 0; break; + case MSR_MTRRcap: + data = 0x500 | KVM_NR_VAR_MTRR; + break; + case 0x200 ... 0x2ff: + return get_msr_mtrr(vcpu, msr, pdata); case 0xcd: /* fsb frequency */ data = 3; break; diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index 075598b4e3f3..fc72bad878ed 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -79,6 +79,7 @@ #define KVM_MIN_FREE_MMU_PAGES 5 #define KVM_REFILL_PAGES 25 #define KVM_MAX_CPUID_ENTRIES 40 +#define KVM_NR_VAR_MTRR 8 extern spinlock_t kvm_lock; extern struct list_head vm_list; @@ -290,6 +291,8 @@ struct kvm_vcpu_arch { struct page *time_page; bool nmi_pending; + + u64 mtrr[0x100]; }; struct kvm_mem_alias { -- cgit From 3e6e0aab1ba1e8b354ce01f5659336f9aee69437 Mon Sep 17 00:00:00 2001 From: Guillaume Thouvenin Date: Tue, 27 May 2008 10:18:46 +0200 Subject: KVM: Prefixes segment functions that will be exported with "kvm_" Prefixes functions that will be exported with kvm_. We also prefixed set_segment() even if it still static to be coherent. signed-off-by: Guillaume Thouvenin Signed-off-by: Laurent Vivier Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 78 +++++++++++++++++++++++----------------------- include/asm-x86/kvm_host.h | 4 +++ 2 files changed, 43 insertions(+), 39 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 5f67a7c54e82..4c94fad7f01e 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3100,8 +3100,8 @@ int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) return 0; } -static void get_segment(struct kvm_vcpu *vcpu, - struct kvm_segment *var, int seg) +void kvm_get_segment(struct kvm_vcpu *vcpu, + struct kvm_segment *var, int seg) { kvm_x86_ops->get_segment(vcpu, var, seg); } @@ -3110,7 +3110,7 @@ void kvm_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) { struct kvm_segment cs; - get_segment(vcpu, &cs, VCPU_SREG_CS); + kvm_get_segment(vcpu, &cs, VCPU_SREG_CS); *db = cs.db; *l = cs.l; } @@ -3124,15 +3124,15 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, vcpu_load(vcpu); - get_segment(vcpu, &sregs->cs, VCPU_SREG_CS); - get_segment(vcpu, &sregs->ds, VCPU_SREG_DS); - get_segment(vcpu, &sregs->es, VCPU_SREG_ES); - get_segment(vcpu, &sregs->fs, VCPU_SREG_FS); - get_segment(vcpu, &sregs->gs, VCPU_SREG_GS); - get_segment(vcpu, &sregs->ss, VCPU_SREG_SS); + kvm_get_segment(vcpu, &sregs->cs, VCPU_SREG_CS); + kvm_get_segment(vcpu, &sregs->ds, VCPU_SREG_DS); + kvm_get_segment(vcpu, &sregs->es, VCPU_SREG_ES); + kvm_get_segment(vcpu, &sregs->fs, VCPU_SREG_FS); + kvm_get_segment(vcpu, &sregs->gs, VCPU_SREG_GS); + kvm_get_segment(vcpu, &sregs->ss, VCPU_SREG_SS); - get_segment(vcpu, &sregs->tr, VCPU_SREG_TR); - get_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); + kvm_get_segment(vcpu, &sregs->tr, VCPU_SREG_TR); + kvm_get_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); kvm_x86_ops->get_idt(vcpu, &dt); sregs->idt.limit = dt.limit; @@ -3184,7 +3184,7 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, return 0; } -static void set_segment(struct kvm_vcpu *vcpu, +static void kvm_set_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg) { kvm_x86_ops->set_segment(vcpu, var, seg); @@ -3221,7 +3221,7 @@ static void get_segment_descritptor_dtable(struct kvm_vcpu *vcpu, if (selector & 1 << 2) { struct kvm_segment kvm_seg; - get_segment(vcpu, &kvm_seg, VCPU_SREG_LDTR); + kvm_get_segment(vcpu, &kvm_seg, VCPU_SREG_LDTR); if (kvm_seg.unusable) dtable->limit = 0; @@ -3327,7 +3327,7 @@ static u16 get_segment_selector(struct kvm_vcpu *vcpu, int seg) { struct kvm_segment kvm_seg; - get_segment(vcpu, &kvm_seg, seg); + kvm_get_segment(vcpu, &kvm_seg, seg); return kvm_seg.selector; } @@ -3343,8 +3343,8 @@ static int load_segment_descriptor_to_kvm_desct(struct kvm_vcpu *vcpu, return 0; } -static int load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, - int type_bits, int seg) +int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, + int type_bits, int seg) { struct kvm_segment kvm_seg; @@ -3357,7 +3357,7 @@ static int load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, if (!kvm_seg.s) kvm_seg.unusable = 1; - set_segment(vcpu, &kvm_seg, seg); + kvm_set_segment(vcpu, &kvm_seg, seg); return 0; } @@ -3403,25 +3403,25 @@ static int load_state_from_tss32(struct kvm_vcpu *vcpu, vcpu->arch.regs[VCPU_REGS_RSI] = tss->esi; vcpu->arch.regs[VCPU_REGS_RDI] = tss->edi; - if (load_segment_descriptor(vcpu, tss->ldt_selector, 0, VCPU_SREG_LDTR)) + if (kvm_load_segment_descriptor(vcpu, tss->ldt_selector, 0, VCPU_SREG_LDTR)) return 1; - if (load_segment_descriptor(vcpu, tss->es, 1, VCPU_SREG_ES)) + if (kvm_load_segment_descriptor(vcpu, tss->es, 1, VCPU_SREG_ES)) return 1; - if (load_segment_descriptor(vcpu, tss->cs, 9, VCPU_SREG_CS)) + if (kvm_load_segment_descriptor(vcpu, tss->cs, 9, VCPU_SREG_CS)) return 1; - if (load_segment_descriptor(vcpu, tss->ss, 1, VCPU_SREG_SS)) + if (kvm_load_segment_descriptor(vcpu, tss->ss, 1, VCPU_SREG_SS)) return 1; - if (load_segment_descriptor(vcpu, tss->ds, 1, VCPU_SREG_DS)) + if (kvm_load_segment_descriptor(vcpu, tss->ds, 1, VCPU_SREG_DS)) return 1; - if (load_segment_descriptor(vcpu, tss->fs, 1, VCPU_SREG_FS)) + if (kvm_load_segment_descriptor(vcpu, tss->fs, 1, VCPU_SREG_FS)) return 1; - if (load_segment_descriptor(vcpu, tss->gs, 1, VCPU_SREG_GS)) + if (kvm_load_segment_descriptor(vcpu, tss->gs, 1, VCPU_SREG_GS)) return 1; return 0; } @@ -3462,19 +3462,19 @@ static int load_state_from_tss16(struct kvm_vcpu *vcpu, vcpu->arch.regs[VCPU_REGS_RSI] = tss->si; vcpu->arch.regs[VCPU_REGS_RDI] = tss->di; - if (load_segment_descriptor(vcpu, tss->ldt, 0, VCPU_SREG_LDTR)) + if (kvm_load_segment_descriptor(vcpu, tss->ldt, 0, VCPU_SREG_LDTR)) return 1; - if (load_segment_descriptor(vcpu, tss->es, 1, VCPU_SREG_ES)) + if (kvm_load_segment_descriptor(vcpu, tss->es, 1, VCPU_SREG_ES)) return 1; - if (load_segment_descriptor(vcpu, tss->cs, 9, VCPU_SREG_CS)) + if (kvm_load_segment_descriptor(vcpu, tss->cs, 9, VCPU_SREG_CS)) return 1; - if (load_segment_descriptor(vcpu, tss->ss, 1, VCPU_SREG_SS)) + if (kvm_load_segment_descriptor(vcpu, tss->ss, 1, VCPU_SREG_SS)) return 1; - if (load_segment_descriptor(vcpu, tss->ds, 1, VCPU_SREG_DS)) + if (kvm_load_segment_descriptor(vcpu, tss->ds, 1, VCPU_SREG_DS)) return 1; return 0; } @@ -3532,7 +3532,7 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) struct desc_struct nseg_desc; int ret = 0; - get_segment(vcpu, &tr_seg, VCPU_SREG_TR); + kvm_get_segment(vcpu, &tr_seg, VCPU_SREG_TR); if (load_guest_segment_descriptor(vcpu, tss_selector, &nseg_desc)) goto out; @@ -3591,7 +3591,7 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason) kvm_x86_ops->set_cr0(vcpu, vcpu->arch.cr0 | X86_CR0_TS); seg_desct_to_kvm_desct(&nseg_desc, tss_selector, &tr_seg); tr_seg.type = 11; - set_segment(vcpu, &tr_seg, VCPU_SREG_TR); + kvm_set_segment(vcpu, &tr_seg, VCPU_SREG_TR); out: kvm_x86_ops->decache_regs(vcpu); return ret; @@ -3658,15 +3658,15 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, } } - set_segment(vcpu, &sregs->cs, VCPU_SREG_CS); - set_segment(vcpu, &sregs->ds, VCPU_SREG_DS); - set_segment(vcpu, &sregs->es, VCPU_SREG_ES); - set_segment(vcpu, &sregs->fs, VCPU_SREG_FS); - set_segment(vcpu, &sregs->gs, VCPU_SREG_GS); - set_segment(vcpu, &sregs->ss, VCPU_SREG_SS); + kvm_set_segment(vcpu, &sregs->cs, VCPU_SREG_CS); + kvm_set_segment(vcpu, &sregs->ds, VCPU_SREG_DS); + kvm_set_segment(vcpu, &sregs->es, VCPU_SREG_ES); + kvm_set_segment(vcpu, &sregs->fs, VCPU_SREG_FS); + kvm_set_segment(vcpu, &sregs->gs, VCPU_SREG_GS); + kvm_set_segment(vcpu, &sregs->ss, VCPU_SREG_SS); - set_segment(vcpu, &sregs->tr, VCPU_SREG_TR); - set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); + kvm_set_segment(vcpu, &sregs->tr, VCPU_SREG_TR); + kvm_set_segment(vcpu, &sregs->ldt, VCPU_SREG_LDTR); vcpu_put(vcpu); diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index fc72bad878ed..cd6a4bb8c8e8 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -503,6 +503,10 @@ int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr, int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value); +void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg); +int kvm_load_segment_descriptor(struct kvm_vcpu *vcpu, u16 selector, + int type_bits, int seg); + int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason); void kvm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0); -- cgit From 92760499d01ef91518119908eb9b8798b6c9bd3f Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Fri, 30 May 2008 16:05:53 +0200 Subject: KVM: kvm_io_device: extend in_range() to manage len and write attribute Modify member in_range() of structure kvm_io_device to pass length and the type of the I/O (write or read). This modification allows to use kvm_io_device with coalesced MMIO. Signed-off-by: Laurent Vivier Signed-off-by: Avi Kivity --- arch/ia64/kvm/kvm-ia64.c | 6 +++--- arch/x86/kvm/i8254.c | 6 ++++-- arch/x86/kvm/i8259.c | 3 ++- arch/x86/kvm/lapic.c | 3 ++- arch/x86/kvm/x86.c | 28 +++++++++++++++++----------- include/linux/kvm_host.h | 3 ++- virt/kvm/ioapic.c | 3 ++- virt/kvm/iodev.h | 8 +++++--- virt/kvm/kvm_main.c | 5 +++-- 9 files changed, 40 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 7c504be57972..bb58df7cc418 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -195,11 +195,11 @@ int kvm_dev_ioctl_check_extension(long ext) } static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu, - gpa_t addr) + gpa_t addr, int len, int is_write) { struct kvm_io_device *dev; - dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr); + dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr, len, is_write); return dev; } @@ -231,7 +231,7 @@ static int handle_mmio(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) kvm_run->exit_reason = KVM_EXIT_MMIO; return 0; mmio: - mmio_dev = vcpu_find_mmio_dev(vcpu, p->addr); + mmio_dev = vcpu_find_mmio_dev(vcpu, p->addr, p->size, !p->dir); if (mmio_dev) { if (!p->dir) kvm_iodevice_write(mmio_dev, p->addr, p->size, diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 60074dc66bd7..9e3391e9a1b7 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -460,7 +460,8 @@ static void pit_ioport_read(struct kvm_io_device *this, mutex_unlock(&pit_state->lock); } -static int pit_in_range(struct kvm_io_device *this, gpa_t addr) +static int pit_in_range(struct kvm_io_device *this, gpa_t addr, + int len, int is_write) { return ((addr >= KVM_PIT_BASE_ADDRESS) && (addr < KVM_PIT_BASE_ADDRESS + KVM_PIT_MEM_LENGTH)); @@ -501,7 +502,8 @@ static void speaker_ioport_read(struct kvm_io_device *this, mutex_unlock(&pit_state->lock); } -static int speaker_in_range(struct kvm_io_device *this, gpa_t addr) +static int speaker_in_range(struct kvm_io_device *this, gpa_t addr, + int len, int is_write) { return (addr == KVM_SPEAKER_BASE_ADDRESS); } diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index ab29cf2def47..5857f59ad4aa 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -346,7 +346,8 @@ static u32 elcr_ioport_read(void *opaque, u32 addr1) return s->elcr; } -static int picdev_in_range(struct kvm_io_device *this, gpa_t addr) +static int picdev_in_range(struct kvm_io_device *this, gpa_t addr, + int len, int is_write) { switch (addr) { case 0x20: diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index e48d19394031..180ba7316da5 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -785,7 +785,8 @@ static void apic_mmio_write(struct kvm_io_device *this, } -static int apic_mmio_range(struct kvm_io_device *this, gpa_t addr) +static int apic_mmio_range(struct kvm_io_device *this, gpa_t addr, + int len, int size) { struct kvm_lapic *apic = (struct kvm_lapic *)this->private; int ret = 0; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 4c94fad7f01e..ab3f5552d694 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1797,13 +1797,14 @@ static void kvm_init_msr_list(void) * Only apic need an MMIO device hook, so shortcut now.. */ static struct kvm_io_device *vcpu_find_pervcpu_dev(struct kvm_vcpu *vcpu, - gpa_t addr) + gpa_t addr, int len, + int is_write) { struct kvm_io_device *dev; if (vcpu->arch.apic) { dev = &vcpu->arch.apic->dev; - if (dev->in_range(dev, addr)) + if (dev->in_range(dev, addr, len, is_write)) return dev; } return NULL; @@ -1811,13 +1812,15 @@ static struct kvm_io_device *vcpu_find_pervcpu_dev(struct kvm_vcpu *vcpu, static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu, - gpa_t addr) + gpa_t addr, int len, + int is_write) { struct kvm_io_device *dev; - dev = vcpu_find_pervcpu_dev(vcpu, addr); + dev = vcpu_find_pervcpu_dev(vcpu, addr, len, is_write); if (dev == NULL) - dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr); + dev = kvm_io_bus_find_dev(&vcpu->kvm->mmio_bus, addr, len, + is_write); return dev; } @@ -1885,7 +1888,7 @@ mmio: * Is this MMIO handled locally? */ mutex_lock(&vcpu->kvm->lock); - mmio_dev = vcpu_find_mmio_dev(vcpu, gpa); + mmio_dev = vcpu_find_mmio_dev(vcpu, gpa, bytes, 0); if (mmio_dev) { kvm_iodevice_read(mmio_dev, gpa, bytes, val); mutex_unlock(&vcpu->kvm->lock); @@ -1940,7 +1943,7 @@ mmio: * Is this MMIO handled locally? */ mutex_lock(&vcpu->kvm->lock); - mmio_dev = vcpu_find_mmio_dev(vcpu, gpa); + mmio_dev = vcpu_find_mmio_dev(vcpu, gpa, bytes, 1); if (mmio_dev) { kvm_iodevice_write(mmio_dev, gpa, bytes, val); mutex_unlock(&vcpu->kvm->lock); @@ -2317,9 +2320,10 @@ static void pio_string_write(struct kvm_io_device *pio_dev, } static struct kvm_io_device *vcpu_find_pio_dev(struct kvm_vcpu *vcpu, - gpa_t addr) + gpa_t addr, int len, + int is_write) { - return kvm_io_bus_find_dev(&vcpu->kvm->pio_bus, addr); + return kvm_io_bus_find_dev(&vcpu->kvm->pio_bus, addr, len, is_write); } int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, @@ -2351,7 +2355,7 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, kvm_x86_ops->skip_emulated_instruction(vcpu); - pio_dev = vcpu_find_pio_dev(vcpu, port); + pio_dev = vcpu_find_pio_dev(vcpu, port, size, !in); if (pio_dev) { kernel_pio(pio_dev, vcpu, vcpu->arch.pio_data); complete_pio(vcpu); @@ -2433,7 +2437,9 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, } } - pio_dev = vcpu_find_pio_dev(vcpu, port); + pio_dev = vcpu_find_pio_dev(vcpu, port, + vcpu->arch.pio.cur_count, + !vcpu->arch.pio.in); if (!vcpu->arch.pio.in) { /* string PIO write */ ret = pio_copy_data(vcpu); diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 865dcbcb891f..499ff0604234 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -52,7 +52,8 @@ struct kvm_io_bus { void kvm_io_bus_init(struct kvm_io_bus *bus); void kvm_io_bus_destroy(struct kvm_io_bus *bus); -struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, gpa_t addr); +struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, + gpa_t addr, int len, int is_write); void kvm_io_bus_register_dev(struct kvm_io_bus *bus, struct kvm_io_device *dev); diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index d0c668c6959e..c0d22870ee9c 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -307,7 +307,8 @@ void kvm_ioapic_update_eoi(struct kvm *kvm, int vector) __kvm_ioapic_update_eoi(ioapic, i); } -static int ioapic_in_range(struct kvm_io_device *this, gpa_t addr) +static int ioapic_in_range(struct kvm_io_device *this, gpa_t addr, + int len, int is_write) { struct kvm_ioapic *ioapic = (struct kvm_ioapic *)this->private; diff --git a/virt/kvm/iodev.h b/virt/kvm/iodev.h index c14e642027b2..55e8846ac3a6 100644 --- a/virt/kvm/iodev.h +++ b/virt/kvm/iodev.h @@ -27,7 +27,8 @@ struct kvm_io_device { gpa_t addr, int len, const void *val); - int (*in_range)(struct kvm_io_device *this, gpa_t addr); + int (*in_range)(struct kvm_io_device *this, gpa_t addr, int len, + int is_write); void (*destructor)(struct kvm_io_device *this); void *private; @@ -49,9 +50,10 @@ static inline void kvm_iodevice_write(struct kvm_io_device *dev, dev->write(dev, addr, len, val); } -static inline int kvm_iodevice_inrange(struct kvm_io_device *dev, gpa_t addr) +static inline int kvm_iodevice_inrange(struct kvm_io_device *dev, + gpa_t addr, int len, int is_write) { - return dev->in_range(dev, addr); + return dev->in_range(dev, addr, len, is_write); } static inline void kvm_iodevice_destructor(struct kvm_io_device *dev) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 83a0e5ce6037..9330fad2b918 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1350,14 +1350,15 @@ void kvm_io_bus_destroy(struct kvm_io_bus *bus) } } -struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, gpa_t addr) +struct kvm_io_device *kvm_io_bus_find_dev(struct kvm_io_bus *bus, + gpa_t addr, int len, int is_write) { int i; for (i = 0; i < bus->dev_count; i++) { struct kvm_io_device *pos = bus->devs[i]; - if (pos->in_range(pos, addr)) + if (pos->in_range(pos, addr, len, is_write)) return pos; } -- cgit From 5f94c1741bdc7a336553122036e8a779e616ccbf Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Fri, 30 May 2008 16:05:54 +0200 Subject: KVM: Add coalesced MMIO support (common part) This patch adds all needed structures to coalesce MMIOs. Until an architecture uses it, it is not compiled. Coalesced MMIO introduces two ioctl() to define where are the MMIO zones that can be coalesced: - KVM_REGISTER_COALESCED_MMIO registers a coalesced MMIO zone. It requests one parameter (struct kvm_coalesced_mmio_zone) which defines a memory area where MMIOs can be coalesced until the next switch to user space. The maximum number of MMIO zones is KVM_COALESCED_MMIO_ZONE_MAX. - KVM_UNREGISTER_COALESCED_MMIO cancels all registered zones inside the given bounds (bounds are also given by struct kvm_coalesced_mmio_zone). The userspace client can check kernel coalesced MMIO availability by asking ioctl(KVM_CHECK_EXTENSION) for the KVM_CAP_COALESCED_MMIO capability. The ioctl() call to KVM_CAP_COALESCED_MMIO will return 0 if not supported, or the page offset where will be stored the ring buffer. The page offset depends on the architecture. After an ioctl(KVM_RUN), the first page of the KVM memory mapped points to a kvm_run structure. The offset given by KVM_CAP_COALESCED_MMIO is an offset to the coalesced MMIO ring expressed in PAGE_SIZE relatively to the address of the start of th kvm_run structure. The MMIO ring buffer is defined by the structure kvm_coalesced_mmio_ring. [akio: fix oops during guest shutdown] Signed-off-by: Laurent Vivier Signed-off-by: Akio Takebe Signed-off-by: Avi Kivity --- include/linux/kvm.h | 29 +++++++++ include/linux/kvm_host.h | 4 ++ virt/kvm/coalesced_mmio.c | 156 ++++++++++++++++++++++++++++++++++++++++++++++ virt/kvm/coalesced_mmio.h | 23 +++++++ virt/kvm/kvm_main.c | 57 +++++++++++++++++ 5 files changed, 269 insertions(+) create mode 100644 virt/kvm/coalesced_mmio.c create mode 100644 virt/kvm/coalesced_mmio.h (limited to 'include') diff --git a/include/linux/kvm.h b/include/linux/kvm.h index a281afeddfbb..1c908ac29c6c 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -173,6 +173,30 @@ struct kvm_run { }; }; +/* for KVM_REGISTER_COALESCED_MMIO / KVM_UNREGISTER_COALESCED_MMIO */ + +struct kvm_coalesced_mmio_zone { + __u64 addr; + __u32 size; + __u32 pad; +}; + +struct kvm_coalesced_mmio { + __u64 phys_addr; + __u32 len; + __u32 pad; + __u8 data[8]; +}; + +struct kvm_coalesced_mmio_ring { + __u32 first, last; + struct kvm_coalesced_mmio coalesced_mmio[0]; +}; + +#define KVM_COALESCED_MMIO_MAX \ + ((PAGE_SIZE - sizeof(struct kvm_coalesced_mmio_ring)) / \ + sizeof(struct kvm_coalesced_mmio)) + /* for KVM_TRANSLATE */ struct kvm_translation { /* in */ @@ -346,6 +370,7 @@ struct kvm_trace_rec { #define KVM_CAP_NOP_IO_DELAY 12 #define KVM_CAP_PV_MMU 13 #define KVM_CAP_MP_STATE 14 +#define KVM_CAP_COALESCED_MMIO 15 /* * ioctls for VM fds @@ -371,6 +396,10 @@ struct kvm_trace_rec { #define KVM_CREATE_PIT _IO(KVMIO, 0x64) #define KVM_GET_PIT _IOWR(KVMIO, 0x65, struct kvm_pit_state) #define KVM_SET_PIT _IOR(KVMIO, 0x66, struct kvm_pit_state) +#define KVM_REGISTER_COALESCED_MMIO \ + _IOW(KVMIO, 0x67, struct kvm_coalesced_mmio_zone) +#define KVM_UNREGISTER_COALESCED_MMIO \ + _IOW(KVMIO, 0x68, struct kvm_coalesced_mmio_zone) /* * ioctls for vcpu fds diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 499ff0604234..d220b4926c4a 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -117,6 +117,10 @@ struct kvm { struct kvm_vm_stat stat; struct kvm_arch arch; atomic_t users_count; +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + struct kvm_coalesced_mmio_dev *coalesced_mmio_dev; + struct kvm_coalesced_mmio_ring *coalesced_mmio_ring; +#endif }; /* The guest did something we don't support. */ diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c new file mode 100644 index 000000000000..5ae620d32fac --- /dev/null +++ b/virt/kvm/coalesced_mmio.c @@ -0,0 +1,156 @@ +/* + * KVM coalesced MMIO + * + * Copyright (c) 2008 Bull S.A.S. + * + * Author: Laurent Vivier + * + */ + +#include "iodev.h" + +#include +#include + +#include "coalesced_mmio.h" + +static int coalesced_mmio_in_range(struct kvm_io_device *this, + gpa_t addr, int len, int is_write) +{ + struct kvm_coalesced_mmio_dev *dev = + (struct kvm_coalesced_mmio_dev*)this->private; + struct kvm_coalesced_mmio_zone *zone; + int next; + int i; + + if (!is_write) + return 0; + + /* kvm->lock is taken by the caller and must be not released before + * dev.read/write + */ + + /* Are we able to batch it ? */ + + /* last is the first free entry + * check if we don't meet the first used entry + * there is always one unused entry in the buffer + */ + + next = (dev->kvm->coalesced_mmio_ring->last + 1) % + KVM_COALESCED_MMIO_MAX; + if (next == dev->kvm->coalesced_mmio_ring->first) { + /* full */ + return 0; + } + + /* is it in a batchable area ? */ + + for (i = 0; i < dev->nb_zones; i++) { + zone = &dev->zone[i]; + + /* (addr,len) is fully included in + * (zone->addr, zone->size) + */ + + if (zone->addr <= addr && + addr + len <= zone->addr + zone->size) + return 1; + } + return 0; +} + +static void coalesced_mmio_write(struct kvm_io_device *this, + gpa_t addr, int len, const void *val) +{ + struct kvm_coalesced_mmio_dev *dev = + (struct kvm_coalesced_mmio_dev*)this->private; + struct kvm_coalesced_mmio_ring *ring = dev->kvm->coalesced_mmio_ring; + + /* kvm->lock must be taken by caller before call to in_range()*/ + + /* copy data in first free entry of the ring */ + + ring->coalesced_mmio[ring->last].phys_addr = addr; + ring->coalesced_mmio[ring->last].len = len; + memcpy(ring->coalesced_mmio[ring->last].data, val, len); + smp_wmb(); + ring->last = (ring->last + 1) % KVM_COALESCED_MMIO_MAX; +} + +static void coalesced_mmio_destructor(struct kvm_io_device *this) +{ + kfree(this); +} + +int kvm_coalesced_mmio_init(struct kvm *kvm) +{ + struct kvm_coalesced_mmio_dev *dev; + + dev = kzalloc(sizeof(struct kvm_coalesced_mmio_dev), GFP_KERNEL); + if (!dev) + return -ENOMEM; + dev->dev.write = coalesced_mmio_write; + dev->dev.in_range = coalesced_mmio_in_range; + dev->dev.destructor = coalesced_mmio_destructor; + dev->dev.private = dev; + dev->kvm = kvm; + kvm->coalesced_mmio_dev = dev; + kvm_io_bus_register_dev(&kvm->mmio_bus, &dev->dev); + + return 0; +} + +int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, + struct kvm_coalesced_mmio_zone *zone) +{ + struct kvm_coalesced_mmio_dev *dev = kvm->coalesced_mmio_dev; + + if (dev == NULL) + return -EINVAL; + + mutex_lock(&kvm->lock); + if (dev->nb_zones >= KVM_COALESCED_MMIO_ZONE_MAX) { + mutex_unlock(&kvm->lock); + return -ENOBUFS; + } + + dev->zone[dev->nb_zones] = *zone; + dev->nb_zones++; + + mutex_unlock(&kvm->lock); + return 0; +} + +int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, + struct kvm_coalesced_mmio_zone *zone) +{ + int i; + struct kvm_coalesced_mmio_dev *dev = kvm->coalesced_mmio_dev; + struct kvm_coalesced_mmio_zone *z; + + if (dev == NULL) + return -EINVAL; + + mutex_lock(&kvm->lock); + + i = dev->nb_zones; + while(i) { + z = &dev->zone[i - 1]; + + /* unregister all zones + * included in (zone->addr, zone->size) + */ + + if (zone->addr <= z->addr && + z->addr + z->size <= zone->addr + zone->size) { + dev->nb_zones--; + *z = dev->zone[dev->nb_zones]; + } + i--; + } + + mutex_unlock(&kvm->lock); + + return 0; +} diff --git a/virt/kvm/coalesced_mmio.h b/virt/kvm/coalesced_mmio.h new file mode 100644 index 000000000000..5ac0ec628461 --- /dev/null +++ b/virt/kvm/coalesced_mmio.h @@ -0,0 +1,23 @@ +/* + * KVM coalesced MMIO + * + * Copyright (c) 2008 Bull S.A.S. + * + * Author: Laurent Vivier + * + */ + +#define KVM_COALESCED_MMIO_ZONE_MAX 100 + +struct kvm_coalesced_mmio_dev { + struct kvm_io_device dev; + struct kvm *kvm; + int nb_zones; + struct kvm_coalesced_mmio_zone zone[KVM_COALESCED_MMIO_ZONE_MAX]; +}; + +int kvm_coalesced_mmio_init(struct kvm *kvm); +int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, + struct kvm_coalesced_mmio_zone *zone); +int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, + struct kvm_coalesced_mmio_zone *zone); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 9330fad2b918..7d10dfa0d388 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -47,6 +47,10 @@ #include #include +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET +#include "coalesced_mmio.h" +#endif + MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); @@ -185,10 +189,23 @@ EXPORT_SYMBOL_GPL(kvm_vcpu_uninit); static struct kvm *kvm_create_vm(void) { struct kvm *kvm = kvm_arch_create_vm(); +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + struct page *page; +#endif if (IS_ERR(kvm)) goto out; +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + page = alloc_page(GFP_KERNEL | __GFP_ZERO); + if (!page) { + kfree(kvm); + return ERR_PTR(-ENOMEM); + } + kvm->coalesced_mmio_ring = + (struct kvm_coalesced_mmio_ring *)page_address(page); +#endif + kvm->mm = current->mm; atomic_inc(&kvm->mm->mm_count); spin_lock_init(&kvm->mmu_lock); @@ -200,6 +217,9 @@ static struct kvm *kvm_create_vm(void) spin_lock(&kvm_lock); list_add(&kvm->vm_list, &vm_list); spin_unlock(&kvm_lock); +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + kvm_coalesced_mmio_init(kvm); +#endif out: return kvm; } @@ -242,6 +262,10 @@ static void kvm_destroy_vm(struct kvm *kvm) spin_unlock(&kvm_lock); kvm_io_bus_destroy(&kvm->pio_bus); kvm_io_bus_destroy(&kvm->mmio_bus); +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + if (kvm->coalesced_mmio_ring != NULL) + free_page((unsigned long)kvm->coalesced_mmio_ring); +#endif kvm_arch_destroy_vm(kvm); mmdrop(mm); } @@ -825,6 +849,10 @@ static int kvm_vcpu_fault(struct vm_area_struct *vma, struct vm_fault *vmf) #ifdef CONFIG_X86 else if (vmf->pgoff == KVM_PIO_PAGE_OFFSET) page = virt_to_page(vcpu->arch.pio_data); +#endif +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + else if (vmf->pgoff == KVM_COALESCED_MMIO_PAGE_OFFSET) + page = virt_to_page(vcpu->kvm->coalesced_mmio_ring); #endif else return VM_FAULT_SIGBUS; @@ -1148,6 +1176,32 @@ static long kvm_vm_ioctl(struct file *filp, goto out; break; } +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + case KVM_REGISTER_COALESCED_MMIO: { + struct kvm_coalesced_mmio_zone zone; + r = -EFAULT; + if (copy_from_user(&zone, argp, sizeof zone)) + goto out; + r = -ENXIO; + r = kvm_vm_ioctl_register_coalesced_mmio(kvm, &zone); + if (r) + goto out; + r = 0; + break; + } + case KVM_UNREGISTER_COALESCED_MMIO: { + struct kvm_coalesced_mmio_zone zone; + r = -EFAULT; + if (copy_from_user(&zone, argp, sizeof zone)) + goto out; + r = -ENXIO; + r = kvm_vm_ioctl_unregister_coalesced_mmio(kvm, &zone); + if (r) + goto out; + r = 0; + break; + } +#endif default: r = kvm_arch_vm_ioctl(filp, ioctl, arg); } @@ -1231,6 +1285,9 @@ static long kvm_dev_ioctl(struct file *filp, r = PAGE_SIZE; /* struct kvm_run */ #ifdef CONFIG_X86 r += PAGE_SIZE; /* pio data page */ +#endif +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + r += PAGE_SIZE; /* coalesced mmio ring page */ #endif break; case KVM_TRACE_ENABLE: -- cgit From 542472b53ea9e0add0ba23976018210191d84754 Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Fri, 30 May 2008 16:05:55 +0200 Subject: KVM: Add coalesced MMIO support (x86 part) This patch enables coalesced MMIO for x86 architecture. It defines KVM_MMIO_PAGE_OFFSET and KVM_CAP_COALESCED_MMIO. It enables the compilation of coalesced_mmio.c. Signed-off-by: Laurent Vivier Signed-off-by: Avi Kivity --- arch/x86/kvm/Makefile | 3 ++- arch/x86/kvm/x86.c | 3 +++ include/asm-x86/kvm_host.h | 1 + 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile index c97d35c218db..d0e940bb6f40 100644 --- a/arch/x86/kvm/Makefile +++ b/arch/x86/kvm/Makefile @@ -2,7 +2,8 @@ # Makefile for Kernel-based Virtual Machine module # -common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o) +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \ + coalesced_mmio.o) ifeq ($(CONFIG_KVM_TRACE),y) common-objs += $(addprefix ../../../virt/kvm/, kvm_trace.o) endif diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ab3f5552d694..d731d4fff1ae 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -885,6 +885,9 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_MP_STATE: r = 1; break; + case KVM_CAP_COALESCED_MMIO: + r = KVM_COALESCED_MMIO_PAGE_OFFSET; + break; case KVM_CAP_VAPIC: r = !kvm_x86_ops->cpu_has_accelerated_tpr(); break; diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index cd6a4bb8c8e8..c64d1242762b 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -27,6 +27,7 @@ #define KVM_PRIVATE_MEM_SLOTS 4 #define KVM_PIO_PAGE_OFFSET 1 +#define KVM_COALESCED_MMIO_PAGE_OFFSET 2 #define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1) #define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD)) -- cgit From 588968b6b7d34e6a88f538d1db9aca47b203623e Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Fri, 30 May 2008 16:05:56 +0200 Subject: KVM: Add coalesced MMIO support (powerpc part) This patch enables coalesced MMIO for powerpc architecture. It defines KVM_MMIO_PAGE_OFFSET and KVM_CAP_COALESCED_MMIO. It enables the compilation of coalesced_mmio.c. Signed-off-by: Laurent Vivier Signed-off-by: Avi Kivity --- arch/powerpc/kvm/Makefile | 2 +- arch/powerpc/kvm/powerpc.c | 3 +++ include/asm-powerpc/kvm_host.h | 2 ++ 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/powerpc/kvm/Makefile b/arch/powerpc/kvm/Makefile index d0d358d367ec..04e3449e1f42 100644 --- a/arch/powerpc/kvm/Makefile +++ b/arch/powerpc/kvm/Makefile @@ -4,7 +4,7 @@ EXTRA_CFLAGS += -Ivirt/kvm -Iarch/powerpc/kvm -common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o coalesced_mmio.o) kvm-objs := $(common-objs) powerpc.o emulate.o booke_guest.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 0513b359851b..b850d2497027 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -145,6 +145,9 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_USER_MEMORY: r = 1; break; + case KVM_CAP_COALESCED_MMIO: + r = KVM_COALESCED_MMIO_PAGE_OFFSET; + break; default: r = 0; break; diff --git a/include/asm-powerpc/kvm_host.h b/include/asm-powerpc/kvm_host.h index 81a69d711017..2655e2a4831e 100644 --- a/include/asm-powerpc/kvm_host.h +++ b/include/asm-powerpc/kvm_host.h @@ -31,6 +31,8 @@ /* memory slots that does not exposed to userspace */ #define KVM_PRIVATE_MEM_SLOTS 4 +#define KVM_COALESCED_MMIO_PAGE_OFFSET 1 + /* We don't currently support large pages. */ #define KVM_PAGES_PER_HPAGE (1<<31) -- cgit From 7f39f8ac177db258200053074aa7a3d98656b1cf Mon Sep 17 00:00:00 2001 From: Laurent Vivier Date: Fri, 30 May 2008 16:05:57 +0200 Subject: KVM: Add coalesced MMIO support (ia64 part) This patch enables coalesced MMIO for ia64 architecture. It defines KVM_MMIO_PAGE_OFFSET and KVM_CAP_COALESCED_MMIO. It enables the compilation of coalesced_mmio.c. [akpm: fix compile error on ia64] Signed-off-by: Laurent Vivier Signed-off-by: Andrew Morton Signed-off-by: Avi Kivity --- arch/ia64/kvm/Makefile | 3 ++- arch/ia64/kvm/kvm-ia64.c | 3 +++ include/asm-ia64/kvm_host.h | 1 + 3 files changed, 6 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/arch/ia64/kvm/Makefile b/arch/ia64/kvm/Makefile index 112791dd2542..bf22fb9e6dcf 100644 --- a/arch/ia64/kvm/Makefile +++ b/arch/ia64/kvm/Makefile @@ -43,7 +43,8 @@ $(obj)/$(offsets-file): arch/ia64/kvm/asm-offsets.s EXTRA_CFLAGS += -Ivirt/kvm -Iarch/ia64/kvm/ EXTRA_AFLAGS += -Ivirt/kvm -Iarch/ia64/kvm/ -common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o) +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \ + coalesced_mmio.o) kvm-objs := $(common-objs) kvm-ia64.o kvm_fw.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index bb58df7cc418..9408b30576d6 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -187,6 +187,9 @@ int kvm_dev_ioctl_check_extension(long ext) r = 1; break; + case KVM_CAP_COALESCED_MMIO: + r = KVM_COALESCED_MMIO_PAGE_OFFSET; + break; default: r = 0; } diff --git a/include/asm-ia64/kvm_host.h b/include/asm-ia64/kvm_host.h index 5c958b0c46b1..1efe513a9941 100644 --- a/include/asm-ia64/kvm_host.h +++ b/include/asm-ia64/kvm_host.h @@ -38,6 +38,7 @@ /* memory slots that does not exposed to userspace */ #define KVM_PRIVATE_MEM_SLOTS 4 +#define KVM_COALESCED_MMIO_PAGE_OFFSET 1 /* define exit reasons from vmm to kvm*/ #define EXIT_REASON_VM_PANIC 0 -- cgit From 9ef621d3be56e1188300476a8102ff54f7b6793f Mon Sep 17 00:00:00 2001 From: "Tan, Li" Date: Fri, 23 May 2008 14:54:09 +0800 Subject: KVM: Support mixed endian machines Currently kvmtrace is not portable. This will prevent from copying a trace file from big-endian target to little-endian workstation for analysis. In the patch, kernel outputs metadata containing a magic number to trace log, and changes 64-bit words to be u64 instead of a pair of u32s. Signed-off-by: Tan Li Acked-by: Jerone Young Acked-by: Hollis Blanchard Signed-off-by: Avi Kivity --- include/linux/kvm.h | 4 ++-- virt/kvm/kvm_trace.c | 18 ++++++++++++------ 2 files changed, 14 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 1c908ac29c6c..0ea064cbfbc8 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -318,14 +318,14 @@ struct kvm_trace_rec { __u32 vcpu_id; union { struct { - __u32 cycle_lo, cycle_hi; + __u64 cycle_u64; __u32 extra_u32[KVM_TRC_EXTRA_MAX]; } cycle; struct { __u32 extra_u32[KVM_TRC_EXTRA_MAX]; } nocycle; } u; -}; +} __attribute__((packed)); #define KVMIO 0xAE diff --git a/virt/kvm/kvm_trace.c b/virt/kvm/kvm_trace.c index 0e495470788d..58141f31ea8f 100644 --- a/virt/kvm/kvm_trace.c +++ b/virt/kvm/kvm_trace.c @@ -72,11 +72,7 @@ static void kvm_add_trace(void *probe_private, void *call_data, rec.cycle_in = p->cycle_in; if (rec.cycle_in) { - u64 cycle = 0; - - cycle = get_cycles(); - rec.u.cycle.cycle_lo = (u32)cycle; - rec.u.cycle.cycle_hi = (u32)(cycle >> 32); + rec.u.cycle.cycle_u64 = get_cycles(); for (i = 0; i < rec.extra_u32; i++) rec.u.cycle.extra_u32[i] = va_arg(*args, u32); @@ -114,8 +110,18 @@ static int kvm_subbuf_start_callback(struct rchan_buf *buf, void *subbuf, { struct kvm_trace *kt; - if (!relay_buf_full(buf)) + if (!relay_buf_full(buf)) { + if (!prev_subbuf) { + /* + * executed only once when the channel is opened + * save metadata as first record + */ + subbuf_start_reserve(buf, sizeof(u32)); + *(u32 *)subbuf = 0x12345678; + } + return 1; + } kt = buf->chan->private_data; atomic_inc(&kt->lost_records); -- cgit From f5b4edcd52e78556800f90d08bfc9126416ac82f Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 15 Jun 2008 22:09:11 -0700 Subject: KVM: x86 emulator: simplify rip relative decoding rip relative decoding is relative to the instruction pointer of the next instruction; by moving address adjustment until after decoding is complete, we remove the need to determine the instruction size. Signed-off-by: Avi Kivity --- arch/x86/kvm/x86_emulate.c | 23 +++++------------------ include/asm-x86/kvm_x86_emulate.h | 1 + 2 files changed, 6 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c index c3a823174f3e..20b604489c3c 100644 --- a/arch/x86/kvm/x86_emulate.c +++ b/arch/x86/kvm/x86_emulate.c @@ -664,7 +664,7 @@ static int decode_modrm(struct x86_emulate_ctxt *ctxt, { struct decode_cache *c = &ctxt->decode; u8 sib; - int index_reg = 0, base_reg = 0, scale, rip_relative = 0; + int index_reg = 0, base_reg = 0, scale; int rc = 0; if (c->rex_prefix) { @@ -754,7 +754,7 @@ static int decode_modrm(struct x86_emulate_ctxt *ctxt, c->modrm_ea += c->regs[index_reg] << scale; } else if ((c->modrm_rm & 7) == 5 && c->modrm_mod == 0) { if (ctxt->mode == X86EMUL_MODE_PROT64) - rip_relative = 1; + c->rip_relative = 1; } else c->modrm_ea += c->regs[c->modrm_rm]; switch (c->modrm_mod) { @@ -770,22 +770,6 @@ static int decode_modrm(struct x86_emulate_ctxt *ctxt, break; } } - if (rip_relative) { - c->modrm_ea += c->eip; - switch (c->d & SrcMask) { - case SrcImmByte: - c->modrm_ea += 1; - break; - case SrcImm: - if (c->d & ByteOp) - c->modrm_ea += 1; - else - if (c->op_bytes == 8) - c->modrm_ea += 4; - else - c->modrm_ea += c->op_bytes; - } - } done: return rc; } @@ -1044,6 +1028,9 @@ done_prefixes: break; } + if (c->rip_relative) + c->modrm_ea += c->eip; + done: return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0; } diff --git a/include/asm-x86/kvm_x86_emulate.h b/include/asm-x86/kvm_x86_emulate.h index b877bbd2d3a7..9fda4b35e195 100644 --- a/include/asm-x86/kvm_x86_emulate.h +++ b/include/asm-x86/kvm_x86_emulate.h @@ -134,6 +134,7 @@ struct decode_cache { u8 modrm_reg; u8 modrm_rm; u8 use_modrm_ea; + bool rip_relative; unsigned long modrm_ea; void *modrm_ptr; unsigned long modrm_val; -- cgit From 7a5b56dfd3a682a51fc84682290d5147872a8e99 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 22 Jun 2008 16:22:51 +0300 Subject: KVM: x86 emulator: lazily evaluate segment registers Instead of prefetching all segment bases before emulation, read them at the last moment. Since most of them are unneeded, we save some cycles on Intel machines where this is a bit expensive. Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 21 --------- arch/x86/kvm/x86_emulate.c | 96 +++++++++++++++++++++++---------------- include/asm-x86/kvm_x86_emulate.h | 10 ++-- 3 files changed, 60 insertions(+), 67 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index d1db5aa5c7f4..f726ba79fd3a 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2126,27 +2126,6 @@ int emulate_instruction(struct kvm_vcpu *vcpu, ? X86EMUL_MODE_PROT64 : cs_db ? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16; - if (vcpu->arch.emulate_ctxt.mode == X86EMUL_MODE_PROT64) { - vcpu->arch.emulate_ctxt.cs_base = 0; - vcpu->arch.emulate_ctxt.ds_base = 0; - vcpu->arch.emulate_ctxt.es_base = 0; - vcpu->arch.emulate_ctxt.ss_base = 0; - } else { - vcpu->arch.emulate_ctxt.cs_base = - get_segment_base(vcpu, VCPU_SREG_CS); - vcpu->arch.emulate_ctxt.ds_base = - get_segment_base(vcpu, VCPU_SREG_DS); - vcpu->arch.emulate_ctxt.es_base = - get_segment_base(vcpu, VCPU_SREG_ES); - vcpu->arch.emulate_ctxt.ss_base = - get_segment_base(vcpu, VCPU_SREG_SS); - } - - vcpu->arch.emulate_ctxt.gs_base = - get_segment_base(vcpu, VCPU_SREG_GS); - vcpu->arch.emulate_ctxt.fs_base = - get_segment_base(vcpu, VCPU_SREG_FS); - r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops); /* Reject the instructions other than VMCALL/VMMCALL when diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c index 38926b7da64a..18ca25c2d4a4 100644 --- a/arch/x86/kvm/x86_emulate.c +++ b/arch/x86/kvm/x86_emulate.c @@ -522,6 +522,39 @@ static inline void jmp_rel(struct decode_cache *c, int rel) register_address_increment(c, &c->eip, rel); } +static void set_seg_override(struct decode_cache *c, int seg) +{ + c->has_seg_override = true; + c->seg_override = seg; +} + +static unsigned long seg_base(struct x86_emulate_ctxt *ctxt, int seg) +{ + if (ctxt->mode == X86EMUL_MODE_PROT64 && seg < VCPU_SREG_FS) + return 0; + + return kvm_x86_ops->get_segment_base(ctxt->vcpu, seg); +} + +static unsigned long seg_override_base(struct x86_emulate_ctxt *ctxt, + struct decode_cache *c) +{ + if (!c->has_seg_override) + return 0; + + return seg_base(ctxt, c->seg_override); +} + +static unsigned long es_base(struct x86_emulate_ctxt *ctxt) +{ + return seg_base(ctxt, VCPU_SREG_ES); +} + +static unsigned long ss_base(struct x86_emulate_ctxt *ctxt) +{ + return seg_base(ctxt, VCPU_SREG_SS); +} + static int do_fetch_insn_byte(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, unsigned long linear, u8 *dest) @@ -735,8 +768,8 @@ static int decode_modrm(struct x86_emulate_ctxt *ctxt, } if (c->modrm_rm == 2 || c->modrm_rm == 3 || (c->modrm_rm == 6 && c->modrm_mod != 0)) - if (!c->override_base) - c->override_base = &ctxt->ss_base; + if (!c->has_seg_override) + set_seg_override(c, VCPU_SREG_SS); c->modrm_ea = (u16)c->modrm_ea; } else { /* 32/64-bit ModR/M decode. */ @@ -807,6 +840,7 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) memset(c, 0, sizeof(struct decode_cache)); c->eip = ctxt->vcpu->arch.rip; + ctxt->cs_base = seg_base(ctxt, VCPU_SREG_CS); memcpy(c->regs, ctxt->vcpu->arch.regs, sizeof c->regs); switch (mode) { @@ -845,23 +879,15 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* switch between 2/4 bytes */ c->ad_bytes = def_ad_bytes ^ 6; break; + case 0x26: /* ES override */ case 0x2e: /* CS override */ - c->override_base = &ctxt->cs_base; - break; + case 0x36: /* SS override */ case 0x3e: /* DS override */ - c->override_base = &ctxt->ds_base; - break; - case 0x26: /* ES override */ - c->override_base = &ctxt->es_base; + set_seg_override(c, (c->b >> 3) & 3); break; case 0x64: /* FS override */ - c->override_base = &ctxt->fs_base; - break; case 0x65: /* GS override */ - c->override_base = &ctxt->gs_base; - break; - case 0x36: /* SS override */ - c->override_base = &ctxt->ss_base; + set_seg_override(c, c->b & 7); break; case 0x40 ... 0x4f: /* REX */ if (mode != X86EMUL_MODE_PROT64) @@ -933,15 +959,11 @@ done_prefixes: if (rc) goto done; - if (!c->override_base) - c->override_base = &ctxt->ds_base; - if (mode == X86EMUL_MODE_PROT64 && - c->override_base != &ctxt->fs_base && - c->override_base != &ctxt->gs_base) - c->override_base = NULL; + if (!c->has_seg_override) + set_seg_override(c, VCPU_SREG_DS); - if (c->override_base && !(!c->twobyte && c->b == 0x8d)) - c->modrm_ea += *c->override_base; + if (!(!c->twobyte && c->b == 0x8d)) + c->modrm_ea += seg_override_base(ctxt, c); if (c->ad_bytes != 8) c->modrm_ea = (u32)c->modrm_ea; @@ -1043,7 +1065,7 @@ static inline void emulate_push(struct x86_emulate_ctxt *ctxt) c->dst.bytes = c->op_bytes; c->dst.val = c->src.val; register_address_increment(c, &c->regs[VCPU_REGS_RSP], -c->op_bytes); - c->dst.ptr = (void *) register_address(c, ctxt->ss_base, + c->dst.ptr = (void *) register_address(c, ss_base(ctxt), c->regs[VCPU_REGS_RSP]); } @@ -1053,7 +1075,7 @@ static inline int emulate_grp1a(struct x86_emulate_ctxt *ctxt, struct decode_cache *c = &ctxt->decode; int rc; - rc = ops->read_std(register_address(c, ctxt->ss_base, + rc = ops->read_std(register_address(c, ss_base(ctxt), c->regs[VCPU_REGS_RSP]), &c->dst.val, c->dst.bytes, ctxt->vcpu); if (rc != 0) @@ -1375,11 +1397,11 @@ special_insn: register_address_increment(c, &c->regs[VCPU_REGS_RSP], -c->op_bytes); c->dst.ptr = (void *) register_address( - c, ctxt->ss_base, c->regs[VCPU_REGS_RSP]); + c, ss_base(ctxt), c->regs[VCPU_REGS_RSP]); break; case 0x58 ... 0x5f: /* pop reg */ pop_instruction: - if ((rc = ops->read_std(register_address(c, ctxt->ss_base, + if ((rc = ops->read_std(register_address(c, ss_base(ctxt), c->regs[VCPU_REGS_RSP]), c->dst.ptr, c->op_bytes, ctxt->vcpu)) != 0) goto done; @@ -1405,7 +1427,7 @@ special_insn: c->rep_prefix ? address_mask(c, c->regs[VCPU_REGS_RCX]) : 1, (ctxt->eflags & EFLG_DF), - register_address(c, ctxt->es_base, + register_address(c, es_base(ctxt), c->regs[VCPU_REGS_RDI]), c->rep_prefix, c->regs[VCPU_REGS_RDX]) == 0) { @@ -1421,9 +1443,8 @@ special_insn: c->rep_prefix ? address_mask(c, c->regs[VCPU_REGS_RCX]) : 1, (ctxt->eflags & EFLG_DF), - register_address(c, c->override_base ? - *c->override_base : - ctxt->ds_base, + register_address(c, + seg_override_base(ctxt, c), c->regs[VCPU_REGS_RSI]), c->rep_prefix, c->regs[VCPU_REGS_RDX]) == 0) { @@ -1559,11 +1580,10 @@ special_insn: c->dst.type = OP_MEM; c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes; c->dst.ptr = (unsigned long *)register_address(c, - ctxt->es_base, + es_base(ctxt), c->regs[VCPU_REGS_RDI]); if ((rc = ops->read_emulated(register_address(c, - c->override_base ? *c->override_base : - ctxt->ds_base, + seg_override_base(ctxt, c), c->regs[VCPU_REGS_RSI]), &c->dst.val, c->dst.bytes, ctxt->vcpu)) != 0) @@ -1579,8 +1599,7 @@ special_insn: c->src.type = OP_NONE; /* Disable writeback. */ c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes; c->src.ptr = (unsigned long *)register_address(c, - c->override_base ? *c->override_base : - ctxt->ds_base, + seg_override_base(ctxt, c), c->regs[VCPU_REGS_RSI]); if ((rc = ops->read_emulated((unsigned long)c->src.ptr, &c->src.val, @@ -1591,7 +1610,7 @@ special_insn: c->dst.type = OP_NONE; /* Disable writeback. */ c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes; c->dst.ptr = (unsigned long *)register_address(c, - ctxt->es_base, + es_base(ctxt), c->regs[VCPU_REGS_RDI]); if ((rc = ops->read_emulated((unsigned long)c->dst.ptr, &c->dst.val, @@ -1615,7 +1634,7 @@ special_insn: c->dst.type = OP_MEM; c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes; c->dst.ptr = (unsigned long *)register_address(c, - ctxt->es_base, + es_base(ctxt), c->regs[VCPU_REGS_RDI]); c->dst.val = c->regs[VCPU_REGS_RAX]; register_address_increment(c, &c->regs[VCPU_REGS_RDI], @@ -1627,8 +1646,7 @@ special_insn: c->dst.bytes = (c->d & ByteOp) ? 1 : c->op_bytes; c->dst.ptr = (unsigned long *)&c->regs[VCPU_REGS_RAX]; if ((rc = ops->read_emulated(register_address(c, - c->override_base ? *c->override_base : - ctxt->ds_base, + seg_override_base(ctxt, c), c->regs[VCPU_REGS_RSI]), &c->dst.val, c->dst.bytes, diff --git a/include/asm-x86/kvm_x86_emulate.h b/include/asm-x86/kvm_x86_emulate.h index 9fda4b35e195..4e8c1e48d91d 100644 --- a/include/asm-x86/kvm_x86_emulate.h +++ b/include/asm-x86/kvm_x86_emulate.h @@ -124,7 +124,8 @@ struct decode_cache { u8 rex_prefix; struct operand src; struct operand dst; - unsigned long *override_base; + bool has_seg_override; + u8 seg_override; unsigned int d; unsigned long regs[NR_VCPU_REGS]; unsigned long eip; @@ -151,12 +152,7 @@ struct x86_emulate_ctxt { /* Emulated execution mode, represented by an X86EMUL_MODE value. */ int mode; - unsigned long cs_base; - unsigned long ds_base; - unsigned long es_base; - unsigned long ss_base; - unsigned long gs_base; - unsigned long fs_base; + u32 cs_base; /* decode cache */ -- cgit From 180c12fb22bd17c7187ae1bce023d24a42b2980c Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Fri, 27 Jun 2008 15:05:40 +0200 Subject: KVM: s390: rename private structures While doing some tests with our lcrash implementation I have seen a naming conflict with prefix_info in kvm_host.h vs. addrconf.h To avoid future conflicts lets rename private definitions in asm/kvm_host.h by adding the kvm_s390 prefix. Signed-off-by: Christian Borntraeger Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/kvm/interrupt.c | 32 ++++++++++++++++---------------- arch/s390/kvm/kvm-s390.c | 3 ++- arch/s390/kvm/priv.c | 2 +- arch/s390/kvm/sigp.c | 20 ++++++++++---------- include/asm-s390/kvm_host.h | 36 ++++++++++++++++++------------------ 5 files changed, 47 insertions(+), 46 deletions(-) (limited to 'include') diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index 84a7fed4cd4e..11230b0db957 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -31,7 +31,7 @@ static int psw_interrupts_disabled(struct kvm_vcpu *vcpu) } static int __interrupt_is_deliverable(struct kvm_vcpu *vcpu, - struct interrupt_info *inti) + struct kvm_s390_interrupt_info *inti) { switch (inti->type) { case KVM_S390_INT_EMERGENCY: @@ -91,7 +91,7 @@ static void __set_cpuflag(struct kvm_vcpu *vcpu, u32 flag) } static void __set_intercept_indicator(struct kvm_vcpu *vcpu, - struct interrupt_info *inti) + struct kvm_s390_interrupt_info *inti) { switch (inti->type) { case KVM_S390_INT_EMERGENCY: @@ -111,7 +111,7 @@ static void __set_intercept_indicator(struct kvm_vcpu *vcpu, } static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, - struct interrupt_info *inti) + struct kvm_s390_interrupt_info *inti) { const unsigned short table[] = { 2, 4, 4, 6 }; int rc, exception = 0; @@ -290,9 +290,9 @@ static int __try_deliver_ckc_interrupt(struct kvm_vcpu *vcpu) int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu) { - struct local_interrupt *li = &vcpu->arch.local_int; - struct float_interrupt *fi = vcpu->arch.local_int.float_int; - struct interrupt_info *inti; + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_float_interrupt *fi = vcpu->arch.local_int.float_int; + struct kvm_s390_interrupt_info *inti; int rc = 0; if (atomic_read(&li->active)) { @@ -408,9 +408,9 @@ void kvm_s390_idle_wakeup(unsigned long data) void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu) { - struct local_interrupt *li = &vcpu->arch.local_int; - struct float_interrupt *fi = vcpu->arch.local_int.float_int; - struct interrupt_info *n, *inti = NULL; + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_float_interrupt *fi = vcpu->arch.local_int.float_int; + struct kvm_s390_interrupt_info *n, *inti = NULL; int deliver; __reset_intercept_indicators(vcpu); @@ -465,8 +465,8 @@ void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu) int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code) { - struct local_interrupt *li = &vcpu->arch.local_int; - struct interrupt_info *inti; + struct kvm_s390_local_interrupt *li = &vcpu->arch.local_int; + struct kvm_s390_interrupt_info *inti; inti = kzalloc(sizeof(*inti), GFP_KERNEL); if (!inti) @@ -487,9 +487,9 @@ int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code) int kvm_s390_inject_vm(struct kvm *kvm, struct kvm_s390_interrupt *s390int) { - struct local_interrupt *li; - struct float_interrupt *fi; - struct interrupt_info *inti; + struct kvm_s390_local_interrupt *li; + struct kvm_s390_float_interrupt *fi; + struct kvm_s390_interrupt_info *inti; int sigcpu; inti = kzalloc(sizeof(*inti), GFP_KERNEL); @@ -544,8 +544,8 @@ int kvm_s390_inject_vm(struct kvm *kvm, int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, struct kvm_s390_interrupt *s390int) { - struct local_interrupt *li; - struct interrupt_info *inti; + struct kvm_s390_local_interrupt *li; + struct kvm_s390_interrupt_info *inti; inti = kzalloc(sizeof(*inti), GFP_KERNEL); if (!inti) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index cdab57c5bc70..399acf3f64dd 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -275,7 +275,8 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, if (!vcpu) goto out_nomem; - vcpu->arch.sie_block = (struct sie_block *) get_zeroed_page(GFP_KERNEL); + vcpu->arch.sie_block = (struct kvm_s390_sie_block *) + get_zeroed_page(GFP_KERNEL); if (!vcpu->arch.sie_block) goto out_free_cpu; diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index c02286c6a931..2e2d2ffb6a07 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -199,7 +199,7 @@ out: static void handle_stsi_3_2_2(struct kvm_vcpu *vcpu, struct sysinfo_3_2_2 *mem) { - struct float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; int cpus = 0; int n; diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c index 0a236acfb5f6..5a556114eaa5 100644 --- a/arch/s390/kvm/sigp.c +++ b/arch/s390/kvm/sigp.c @@ -45,7 +45,7 @@ static int __sigp_sense(struct kvm_vcpu *vcpu, u16 cpu_addr, u64 *reg) { - struct float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; int rc; if (cpu_addr >= KVM_MAX_VCPUS) @@ -71,9 +71,9 @@ static int __sigp_sense(struct kvm_vcpu *vcpu, u16 cpu_addr, u64 *reg) static int __sigp_emergency(struct kvm_vcpu *vcpu, u16 cpu_addr) { - struct float_interrupt *fi = &vcpu->kvm->arch.float_int; - struct local_interrupt *li; - struct interrupt_info *inti; + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_local_interrupt *li; + struct kvm_s390_interrupt_info *inti; int rc; if (cpu_addr >= KVM_MAX_VCPUS) @@ -108,9 +108,9 @@ unlock: static int __sigp_stop(struct kvm_vcpu *vcpu, u16 cpu_addr, int store) { - struct float_interrupt *fi = &vcpu->kvm->arch.float_int; - struct local_interrupt *li; - struct interrupt_info *inti; + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_local_interrupt *li; + struct kvm_s390_interrupt_info *inti; int rc; if (cpu_addr >= KVM_MAX_VCPUS) @@ -169,9 +169,9 @@ static int __sigp_set_arch(struct kvm_vcpu *vcpu, u32 parameter) static int __sigp_set_prefix(struct kvm_vcpu *vcpu, u16 cpu_addr, u32 address, u64 *reg) { - struct float_interrupt *fi = &vcpu->kvm->arch.float_int; - struct local_interrupt *li; - struct interrupt_info *inti; + struct kvm_s390_float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct kvm_s390_local_interrupt *li; + struct kvm_s390_interrupt_info *inti; int rc; u8 tmp; diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h index 18cbd8a39796..3234dd5b3511 100644 --- a/include/asm-s390/kvm_host.h +++ b/include/asm-s390/kvm_host.h @@ -62,7 +62,7 @@ struct sca_block { #define CPUSTAT_J 0x00000002 #define CPUSTAT_P 0x00000001 -struct sie_block { +struct kvm_s390_sie_block { atomic_t cpuflags; /* 0x0000 */ __u32 prefix; /* 0x0004 */ __u8 reserved8[32]; /* 0x0008 */ @@ -140,14 +140,14 @@ struct kvm_vcpu_stat { u32 diagnose_44; }; -struct io_info { +struct kvm_s390_io_info { __u16 subchannel_id; /* 0x0b8 */ __u16 subchannel_nr; /* 0x0ba */ __u32 io_int_parm; /* 0x0bc */ __u32 io_int_word; /* 0x0c0 */ }; -struct ext_info { +struct kvm_s390_ext_info { __u32 ext_params; __u64 ext_params2; }; @@ -160,22 +160,22 @@ struct ext_info { #define PGM_SPECIFICATION 0x06 #define PGM_DATA 0x07 -struct pgm_info { +struct kvm_s390_pgm_info { __u16 code; }; -struct prefix_info { +struct kvm_s390_prefix_info { __u32 address; }; -struct interrupt_info { +struct kvm_s390_interrupt_info { struct list_head list; u64 type; union { - struct io_info io; - struct ext_info ext; - struct pgm_info pgm; - struct prefix_info prefix; + struct kvm_s390_io_info io; + struct kvm_s390_ext_info ext; + struct kvm_s390_pgm_info pgm; + struct kvm_s390_prefix_info prefix; }; }; @@ -183,35 +183,35 @@ struct interrupt_info { #define ACTION_STORE_ON_STOP 1 #define ACTION_STOP_ON_STOP 2 -struct local_interrupt { +struct kvm_s390_local_interrupt { spinlock_t lock; struct list_head list; atomic_t active; - struct float_interrupt *float_int; + struct kvm_s390_float_interrupt *float_int; int timer_due; /* event indicator for waitqueue below */ wait_queue_head_t wq; atomic_t *cpuflags; unsigned int action_bits; }; -struct float_interrupt { +struct kvm_s390_float_interrupt { spinlock_t lock; struct list_head list; atomic_t active; int next_rr_cpu; unsigned long idle_mask [(64 + sizeof(long) - 1) / sizeof(long)]; - struct local_interrupt *local_int[64]; + struct kvm_s390_local_interrupt *local_int[64]; }; struct kvm_vcpu_arch { - struct sie_block *sie_block; + struct kvm_s390_sie_block *sie_block; unsigned long guest_gprs[16]; s390_fp_regs host_fpregs; unsigned int host_acrs[NUM_ACRS]; s390_fp_regs guest_fpregs; unsigned int guest_acrs[NUM_ACRS]; - struct local_interrupt local_int; + struct kvm_s390_local_interrupt local_int; struct timer_list ckc_timer; union { cpuid_t cpu_id; @@ -228,8 +228,8 @@ struct kvm_arch{ unsigned long guest_memsize; struct sca_block *sca; debug_info_t *dbf; - struct float_interrupt float_int; + struct kvm_s390_float_interrupt float_int; }; -extern int sie64a(struct sie_block *, __u64 *); +extern int sie64a(struct kvm_s390_sie_block *, __u64 *); #endif -- cgit From d6e88aec07aa8f6c7e4024f5734ec659fd7c5a40 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 10 Jul 2008 16:53:33 +0300 Subject: KVM: Prefix some x86 low level function with kvm_, to avoid namespace issues Fixes compilation with CONFIG_VMI enabled. Signed-off-by: Avi Kivity --- arch/x86/kvm/svm.c | 12 ++++++------ arch/x86/kvm/vmx.c | 24 ++++++++++++------------ arch/x86/kvm/x86.c | 18 +++++++++--------- include/asm-x86/kvm_host.h | 26 ++++++++++++-------------- 4 files changed, 39 insertions(+), 41 deletions(-) (limited to 'include') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 858e29702232..b756e876dce3 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1710,9 +1710,9 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) sync_lapic_to_cr8(vcpu); save_host_msrs(vcpu); - fs_selector = read_fs(); - gs_selector = read_gs(); - ldt_selector = read_ldt(); + fs_selector = kvm_read_fs(); + gs_selector = kvm_read_gs(); + ldt_selector = kvm_read_ldt(); svm->host_cr2 = kvm_read_cr2(); svm->host_dr6 = read_dr6(); svm->host_dr7 = read_dr7(); @@ -1845,9 +1845,9 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) write_dr7(svm->host_dr7); kvm_write_cr2(svm->host_cr2); - load_fs(fs_selector); - load_gs(gs_selector); - load_ldt(ldt_selector); + kvm_load_fs(fs_selector); + kvm_load_gs(gs_selector); + kvm_load_ldt(ldt_selector); load_host_msrs(vcpu); reload_tss(vcpu); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index fff3b490976e..0cac63701719 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -484,7 +484,7 @@ static void reload_tss(void) struct descriptor_table gdt; struct desc_struct *descs; - get_gdt(&gdt); + kvm_get_gdt(&gdt); descs = (void *)gdt.base; descs[GDT_ENTRY_TSS].type = 9; /* available TSS */ load_TR_desc(); @@ -540,9 +540,9 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu) * Set host fs and gs selectors. Unfortunately, 22.2.3 does not * allow segment selectors with cpl > 0 or ti == 1. */ - vmx->host_state.ldt_sel = read_ldt(); + vmx->host_state.ldt_sel = kvm_read_ldt(); vmx->host_state.gs_ldt_reload_needed = vmx->host_state.ldt_sel; - vmx->host_state.fs_sel = read_fs(); + vmx->host_state.fs_sel = kvm_read_fs(); if (!(vmx->host_state.fs_sel & 7)) { vmcs_write16(HOST_FS_SELECTOR, vmx->host_state.fs_sel); vmx->host_state.fs_reload_needed = 0; @@ -550,7 +550,7 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu) vmcs_write16(HOST_FS_SELECTOR, 0); vmx->host_state.fs_reload_needed = 1; } - vmx->host_state.gs_sel = read_gs(); + vmx->host_state.gs_sel = kvm_read_gs(); if (!(vmx->host_state.gs_sel & 7)) vmcs_write16(HOST_GS_SELECTOR, vmx->host_state.gs_sel); else { @@ -586,15 +586,15 @@ static void __vmx_load_host_state(struct vcpu_vmx *vmx) ++vmx->vcpu.stat.host_state_reload; vmx->host_state.loaded = 0; if (vmx->host_state.fs_reload_needed) - load_fs(vmx->host_state.fs_sel); + kvm_load_fs(vmx->host_state.fs_sel); if (vmx->host_state.gs_ldt_reload_needed) { - load_ldt(vmx->host_state.ldt_sel); + kvm_load_ldt(vmx->host_state.ldt_sel); /* * If we have to reload gs, we must take care to * preserve our gs base. */ local_irq_save(flags); - load_gs(vmx->host_state.gs_sel); + kvm_load_gs(vmx->host_state.gs_sel); #ifdef CONFIG_X86_64 wrmsrl(MSR_GS_BASE, vmcs_readl(HOST_GS_BASE)); #endif @@ -654,8 +654,8 @@ static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu) * Linux uses per-cpu TSS and GDT, so set these when switching * processors. */ - vmcs_writel(HOST_TR_BASE, read_tr_base()); /* 22.2.4 */ - get_gdt(&dt); + vmcs_writel(HOST_TR_BASE, kvm_read_tr_base()); /* 22.2.4 */ + kvm_get_gdt(&dt); vmcs_writel(HOST_GDTR_BASE, dt.base); /* 22.2.4 */ rdmsrl(MSR_IA32_SYSENTER_ESP, sysenter_esp); @@ -1943,8 +1943,8 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) vmcs_write16(HOST_CS_SELECTOR, __KERNEL_CS); /* 22.2.4 */ vmcs_write16(HOST_DS_SELECTOR, __KERNEL_DS); /* 22.2.4 */ vmcs_write16(HOST_ES_SELECTOR, __KERNEL_DS); /* 22.2.4 */ - vmcs_write16(HOST_FS_SELECTOR, read_fs()); /* 22.2.4 */ - vmcs_write16(HOST_GS_SELECTOR, read_gs()); /* 22.2.4 */ + vmcs_write16(HOST_FS_SELECTOR, kvm_read_fs()); /* 22.2.4 */ + vmcs_write16(HOST_GS_SELECTOR, kvm_read_gs()); /* 22.2.4 */ vmcs_write16(HOST_SS_SELECTOR, __KERNEL_DS); /* 22.2.4 */ #ifdef CONFIG_X86_64 rdmsrl(MSR_FS_BASE, a); @@ -1958,7 +1958,7 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) vmcs_write16(HOST_TR_SELECTOR, GDT_ENTRY_TSS*8); /* 22.2.4 */ - get_idt(&dt); + kvm_get_idt(&dt); vmcs_writel(HOST_IDTR_BASE, dt.base); /* 22.2.4 */ asm("mov $.Lkvm_vmx_return, %0" : "=r"(kvm_vmx_return)); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 89fc8565edee..b131f3c0cf64 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3767,14 +3767,14 @@ void fx_init(struct kvm_vcpu *vcpu) * allocate ram with GFP_KERNEL. */ if (!used_math()) - fx_save(&vcpu->arch.host_fx_image); + kvm_fx_save(&vcpu->arch.host_fx_image); /* Initialize guest FPU by resetting ours and saving into guest's */ preempt_disable(); - fx_save(&vcpu->arch.host_fx_image); - fx_finit(); - fx_save(&vcpu->arch.guest_fx_image); - fx_restore(&vcpu->arch.host_fx_image); + kvm_fx_save(&vcpu->arch.host_fx_image); + kvm_fx_finit(); + kvm_fx_save(&vcpu->arch.guest_fx_image); + kvm_fx_restore(&vcpu->arch.host_fx_image); preempt_enable(); vcpu->arch.cr0 |= X86_CR0_ET; @@ -3791,8 +3791,8 @@ void kvm_load_guest_fpu(struct kvm_vcpu *vcpu) return; vcpu->guest_fpu_loaded = 1; - fx_save(&vcpu->arch.host_fx_image); - fx_restore(&vcpu->arch.guest_fx_image); + kvm_fx_save(&vcpu->arch.host_fx_image); + kvm_fx_restore(&vcpu->arch.guest_fx_image); } EXPORT_SYMBOL_GPL(kvm_load_guest_fpu); @@ -3802,8 +3802,8 @@ void kvm_put_guest_fpu(struct kvm_vcpu *vcpu) return; vcpu->guest_fpu_loaded = 0; - fx_save(&vcpu->arch.guest_fx_image); - fx_restore(&vcpu->arch.host_fx_image); + kvm_fx_save(&vcpu->arch.guest_fx_image); + kvm_fx_restore(&vcpu->arch.host_fx_image); ++vcpu->stat.fpu_reload; } EXPORT_SYMBOL_GPL(kvm_put_guest_fpu); diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index c64d1242762b..f995783b1fdb 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -567,55 +567,53 @@ static inline struct kvm_mmu_page *page_header(hpa_t shadow_page) return (struct kvm_mmu_page *)page_private(page); } -static inline u16 read_fs(void) +static inline u16 kvm_read_fs(void) { u16 seg; asm("mov %%fs, %0" : "=g"(seg)); return seg; } -static inline u16 read_gs(void) +static inline u16 kvm_read_gs(void) { u16 seg; asm("mov %%gs, %0" : "=g"(seg)); return seg; } -static inline u16 read_ldt(void) +static inline u16 kvm_read_ldt(void) { u16 ldt; asm("sldt %0" : "=g"(ldt)); return ldt; } -static inline void load_fs(u16 sel) +static inline void kvm_load_fs(u16 sel) { asm("mov %0, %%fs" : : "rm"(sel)); } -static inline void load_gs(u16 sel) +static inline void kvm_load_gs(u16 sel) { asm("mov %0, %%gs" : : "rm"(sel)); } -#ifndef load_ldt -static inline void load_ldt(u16 sel) +static inline void kvm_load_ldt(u16 sel) { asm("lldt %0" : : "rm"(sel)); } -#endif -static inline void get_idt(struct descriptor_table *table) +static inline void kvm_get_idt(struct descriptor_table *table) { asm("sidt %0" : "=m"(*table)); } -static inline void get_gdt(struct descriptor_table *table) +static inline void kvm_get_gdt(struct descriptor_table *table) { asm("sgdt %0" : "=m"(*table)); } -static inline unsigned long read_tr_base(void) +static inline unsigned long kvm_read_tr_base(void) { u16 tr; asm("str %0" : "=g"(tr)); @@ -632,17 +630,17 @@ static inline unsigned long read_msr(unsigned long msr) } #endif -static inline void fx_save(struct i387_fxsave_struct *image) +static inline void kvm_fx_save(struct i387_fxsave_struct *image) { asm("fxsave (%0)":: "r" (image)); } -static inline void fx_restore(struct i387_fxsave_struct *image) +static inline void kvm_fx_restore(struct i387_fxsave_struct *image) { asm("fxrstor (%0)":: "r" (image)); } -static inline void fx_finit(void) +static inline void kvm_fx_finit(void) { asm("finit"); } -- cgit From 34d4cb8fca1f2a31be152b74797e6cd160ec9de6 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Thu, 10 Jul 2008 20:49:31 -0300 Subject: KVM: MMU: nuke shadowed pgtable pages and ptes on memslot destruction Flush the shadow mmu before removing regions to avoid stale entries. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/ia64/kvm/kvm-ia64.c | 3 +++ arch/powerpc/kvm/powerpc.c | 4 ++++ arch/s390/kvm/kvm-s390.c | 4 ++++ arch/x86/kvm/x86.c | 5 +++++ include/linux/kvm_host.h | 1 + virt/kvm/kvm_main.c | 3 +++ 6 files changed, 20 insertions(+) (limited to 'include') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 9408b30576d6..2672f4d278ac 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1455,6 +1455,9 @@ int kvm_arch_set_memory_region(struct kvm *kvm, return 0; } +void kvm_arch_flush_shadow(struct kvm *kvm) +{ +} long kvm_arch_dev_ioctl(struct file *filp, unsigned int ioctl, unsigned long arg) diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index b850d2497027..53826a5f6c06 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -170,6 +170,10 @@ int kvm_arch_set_memory_region(struct kvm *kvm, return 0; } +void kvm_arch_flush_shadow(struct kvm *kvm) +{ +} + struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id) { struct kvm_vcpu *vcpu; diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 399acf3f64dd..1782cbcd2829 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -675,6 +675,10 @@ int kvm_arch_set_memory_region(struct kvm *kvm, return 0; } +void kvm_arch_flush_shadow(struct kvm *kvm) +{ +} + gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) { return gfn; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b131f3c0cf64..9f1cdb011cff 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4032,6 +4032,11 @@ int kvm_arch_set_memory_region(struct kvm *kvm, return 0; } +void kvm_arch_flush_shadow(struct kvm *kvm) +{ + kvm_mmu_zap_all(kvm); +} + int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) { return vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index d220b4926c4a..07d68a8ae8e9 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -168,6 +168,7 @@ int kvm_arch_set_memory_region(struct kvm *kvm, struct kvm_userspace_memory_region *mem, struct kvm_memory_slot old, int user_alloc); +void kvm_arch_flush_shadow(struct kvm *kvm); gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn); struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn); unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 9ccaf8f5402e..30b36368fcdf 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -405,6 +405,9 @@ int __kvm_set_memory_region(struct kvm *kvm, if (mem->slot >= kvm->nmemslots) kvm->nmemslots = mem->slot + 1; + if (!npages) + kvm_arch_flush_shadow(kvm); + *memslot = new; r = kvm_arch_set_memory_region(kvm, mem, old, user_alloc); -- cgit